- Email: [email protected]
Prospective analysis of complications 30 days after outpatient upper endoscopy1
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 1999 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 94, No. 6, 1999 ISSN 0002-9270/99/$20.00 PII S0002-9270(99)00197-5
Prospective Analysis of Complications 30 Days After Outpatient Upper Endoscopy Richard Zubarik, M.D., Glenn Eisen, M.D., M.P.H., Chris Mastropietro, Jennifer Lopez, John Carroll, M.D., Stanley Benjamin, M.D., and David E. Fleischer, M.D. Department of Gastroenterology, Georgetown University Hospital, Washington, DC, and Gastroenterology Section, VA Medical Center, Memphis, Tennessee
OBJECTIVE: The aim of this study was to compare complication rates reported by patients 30 days after outpatient upper endoscopy with those discussed at our monthly morbidity and mortality conference. We also intended to establish which complications were reported most frequently 30 days after upper endoscopy, and which patients or procedures involved the highest risk. METHODS: Trained interviewers performed standardized telephone interviews on consecutive outpatients undergoing upper endoscopy over a 1-yr period. Patients were queried regarding potential events related to their upper endoscopy in the 30 days subsequent, including symptoms, emergency room (ER) and/or physician visits, and hospitalizations. The indications, findings, and therapies were reviewed from endoscopic reports. RESULTS: A total of 473 patients were contacted 30 days after outpatient upper endoscopy and agreed to participate in our study. The most common complications reported by patients at 30 days were sore throat (9.5%) and abdominal discomfort (5.3%). Twelve patients (2.5%) required an ER/ physician visit and five patients (1.1%) required hospitalization. The minority of both ER/physician visits (16.7%) and hospitalizations (40%) were discussed at our monthly morbidity and mortality conferences. CONCLUSIONS: More complications were reported by patients 30 days after outpatient upper endoscopy than were discussed at our monthly morbidity and mortality conferences. The most frequent complications reported by patients were sore throat and abdominal pain. The minority of ER/ physician visits and hospitalizations were discussed at our morbidity and mortality conferences. (Am J Gastroenterol 1999;94:1539 –1545. © 1999 by Am. Coll. of Gastroenterology)
members regarding endoscopic complications (1). The rate of complications after esophagogastroduodenoscopy (EGD) was 1.3 complications/1000 procedures. Data from this study were derived retrospectively and relied on physician recall or procedural reports. Only 64% of those surveyed returned questionnaires. A similar survey study was performed by the Texas Society for Gastrointestinal Endoscopy in 1979 (2). In this study, the rate of complications after EGD was 2.4 complications/1000 procedures. These studies are important, but they have limitations and may not accurately reflect complications that occur. A prospective study performed at our institution in 1990 evaluated complications discussed at morbidity and mortality conferences over a 1-yr period (3). The complication rate for endoscopic procedures performed was 1.9%. This study included all endoscopic procedures. Patients also were not systematically contacted after their procedure, so all complications may not have been identified. Ambiguity regarding the definition and gradation of complications is also a significant problem (4). Terms such as “major” and “minor” mean different things to different people, thereby making it difficult to compare complications from one study to the next. We used a 30-day follow-up period for procedure-related complications because it is similar to that used for surgical mortality rates (5) and because it has been used in the past to evaluate ERCPrelated complications (6). Complications were graded by recording whether they required an emergency room (ER) and/or physician visit, required hospitalization, or resulted in mortality. The primary question that this study addressed was whether complications were more frequently reported by patients 30 days after outpatient upper endoscopy than were discussed at our monthly morbidity and mortality conferences. We also attempted to determine the most common complications that patients reported at 30 days, and to see if any factors predicted higher risk.
INTRODUCTION There are surprisingly few studies analyzing the complications of upper endoscopy. In 1974, the American Society for Gastrointestinal Endoscopy performed a survey study of its This work was performed at Georgetown University Hospital, Washington, DC.
MATERIALS AND METHODS The Georgetown Upper Endoscopy Cohort (GUEC) consisted of patients undergoing outpatient upper endoscopy at Georgetown University Hospital from January 14, 1997, to January 13, 1998. Sixteen gastroenterologists had patients
1540
Zubarik et al.
AJG – Vol. 94, No. 6, 1999
Table 1. Complications at 30-Day Follow-Up Compared With Those Discussed at Morbidity and Mortality Conference
30-Day follow-up Morbidity and mortality 30-Day follow-up or morbidity and mortality
Procedures
Total Complications (%)
ER/Physician Visits (%)
Hospitalizations (%)
473 473 473
86 (18.2) 10 (2.1) 91 (19.2)
12 (2.5) 2 (0.4) 12 (2.5)
5 (1.1) 2 (0.4) 5 (1.1)
enrolled in our study. Nine were adult academic gastroenterologists, five were private gastroenterologists, and two were pediatric gastroenterologists. The majority of procedures were performed by adult academic gastroenterologists with or without the assistance of a gastroenterology fellow. Three trained interviewers performed standardized telephone interviews on consecutive outpatients undergoing upper endoscopy. Patients were contacted by telephone 30 days after their procedure. If a patient was not contacted on the initial attempt, six repeat attempts were made before excluding the patient from our study. The specific questions that patients were asked were: 1) Did you have any problems that you felt were related to the procedure? If yes, how long did this problem persist? 2) Have you gone to see a doctor or to an emergency room for this problem? 3) Have you been hospitalized for this problem? If yes, why? Patients were excluded if they underwent concomitant colonoscopy, flexible sigmoidoscopy, gastrostomy tube placement, enteroscopy, endoscopic ultrasound, or endoscopic retrograde cholangiopancreatography. Patients were also excluded if they could not speak English or if they did not have a telephone. The indications, findings, and therapies performed during upper endoscopy were reviewed from the endoscopic reports of those patients included in our study. Charts were reviewed of those patients requiring an ER/physician visit or hospitalization. Patients that required hospitalization were discussed with the attending physician who performed the involved procedure. Scheduled hospitalizations (i.e., after esophageal stent placement), or hospitalizations resulting from findings during a procedure (but not a complication) were excluded. Morbidity and Mortality Conferences Both physicians and nurses record endoscopic complications at Georgetown University. Physicians enter complications into their computer-generated endoscopic reports. This is done at the time of the procedure if the complication is immediate or later if it is delayed. Endoscopic reports are created either by the attending physician performing the procedure or by a gastroenterology fellow if one is involved. Nurses enter complications into an endoscopic log book in which every procedure is recorded. The complications reported by either physicians or nurses are then discussed at our monthly morbidity and mortality conferences at which attending physicians, fellows, residents, and nurses are present. All complications that are recorded either in the computerized reports or in the unit log book are discussed at our monthly morbidity and mortality conferences. Compli-
cations occurring during the procedure or afterward are discussed. Discussion of these complications occurs even if the involved attending physician does not come to the meeting. Definition A complication was defined as any deviation from the optimal course after upper endoscopy associated with the procedure as reported by a physician, nurse, or patient. We stratified the severity of reported complications by determining whether they required an ER/physician visit, required a hospitalization, or resulted in mortality.
RESULTS Of 627 patients undergoing outpatient upper endoscopy, 473 (75.4%) were successfully contacted and agreed to participate in our study. The average age of participating patients was 52.7 yr and 54.6% were female. The most common reason for exclusion from our study was inability to contact the patient by telephone; this accounted for 71% of patients who were excluded. This group included patients who did not answer (53 patients), who gave a wrong telephone number (20 patients), whose telephone number was disconnected (18 patients), who were out of town (13 patients), and who did not give a telephone number at all (6 patients). Seven patients who were excluded died before 30-day follow-up. None of these patients died as a result of endoscopy. Ninety-one patients had a complication that was either detected at 30-day follow-up or discussed at our morbidity and mortality conferences (Table 1). Eighty-six patients (18.2%) reported procedure-related problems at 30-day follow-up, and 10 patients (2.1%) were discussed at our morbidity and mortality conferences. Twelve patients required an ER/physician visit as a result of their procedure (2.5%), and five patients required hospitalization (1.1%). There was no procedure-related mortality. Only 16.7% (two of 12) of the patients requiring an ER/physician visit, and 40% (two of five) of the patients requiring hospitalization detected at 30-day follow-up were discussed at morbidity and mortality conferences. Three of the five hospitalized patients had undergone dilation, and were hospitalized to rule out perforation. Two of these patients had fever and one had chest pain (Table 2). One of the hospitalized patients who had undergone dilation had migration of an esophageal stent and subsequently required repeat dilation, stent removal, and placement of another stent. Of the other two patients hospitalized after esophageal
AJG – June, 1999
Complications After Upper Endoscopy
1541
Table 2. Patients Who Required an ER/Physician Visit or Hospitalization After Outpatient Upper Endoscopy Known by Endoscopist
ER/Physician Visit
Hospitalization
Surgery
chest pain
yes
yes
yes
no
yes
fever
yes
yes
yes
no
yes
yes
yes
yes
yes
no
abdominal distention fever
yes
yes
yes
no
none dilation
no no
presyncope torn uvula
no yes
yes yes
yes no
no no
none
no
sore throat
no
yes
no
no
none none
no no
sore throat sore throat
no no
yes yes
no no
no no
none
no
no
yes
no
no
none
no
no
yes
no
no
none
no
abdominal pain abdominal pain chest pain
no
yes
no
no
Patient
Finding
Therapy
1
migrated stent esophageal stricture esophagitis
dilation/stent replacement dilation none
anastomotic stricture gastritis esophageal stricture radiation esophagitis normal cicatricial pemphigoid gastritis
dilation
gastric polyps barrett mucosa
2 3 4 5 6 7 8 9 10 11 12
Reported at M&M no
30-Day Complication
M&M 5 morbidity and mortality conference.
dilation, one had balloon dilation and the other had dilation of a benign stricture. Only one of the three had a documented perforation on radiological imaging (the benign stricture). In this patient the perforation was confined, and did not require surgery. Two patients hospitalized after outpatient upper endoscopy had not undergone esophageal dilation. One of these patients was admitted for presyncope and the other was admitted for a small bowel obstruction. The small bowel obstruction began directly after the procedure in a patient with AIDS, and required surgical intervention the next day after worsening symptoms. The initial upper endoscopy revealed only esophagitis and, at the time of surgery, only adhesions were found.
Thirty-Day Follow-Up A total of 86 patients (18.2%) complained of 95 complications 30 days after outpatient upper endoscopy (Table 3). Oropharyngeal problems (48 patients) and abdominal discomfort (22 patients) were the most common complications reported. Oropharyngeal problems accounted for the most ER/physician visits (four patients), and fever resulted in the most hospitalizations (two patients). Women were more likely to report complications at 30 days than were men (22.5% vs 13%, p , 0.001, x2 test). There was no significant gender difference in ER/physician visits or hospitalizations (Fig. 1). No specific indications, findings, or therapies predicted a significantly higher rate of total complications at
Table 3. Complications 30 Days After Outpatient Upper Endoscopy
Complication
Complications at 30-Day Follow-Up
ER/Physician Visits
Hospitalizations
Oropharyngeal complications (total) Sore throat Sore lip Torn uvula Hoarse Abdominal discomfort (total) Abdominal pain Abdominal distention Nausea/Vomiting Chest Pain Fever/Chills Fatigue Presyncope Other Total
48 45 1 1 1 25 21 4 7 3 3 2 2 5 95
4 3 0 1 0 3 2 1 0 2 2 0 1 0 12
0 0 0 0 0 1 0 1 0 1 2 0 1 0 5
1542
Zubarik et al.
AJG – Vol. 94, No. 6, 1999
Figure 1. Complications by gender.
thirty days. Patients having upper endoscopy performed for the indication of dyspepsia, who had a normal exam, had a higher rate of total reported complications at 30 days (25.5%). This difference was not statistically significant, however, and none of these patients required an ER/physician visit or hospitalization. Reported total complication rates at 30 days were slightly higher for therapeutic examinations than for diagnostic examinations (19.6% vs 17.3%). This was also true for ER/physician visits and hospitalizations. When diagnostic procedures were performed, 2.2% of patients required an ER/physician visit and 0.5% of patients required hospitalization (Table 4). When therapeutic exams were performed, 4.1% of patients required an ER/physician visit, and 3.1% required hospitalization. The therapy resulting in the most hospitalizations was endoscopic dilation, which had an overall hospitalization rate of 5%. None of the diagnostic exams resulted in perforation. There was one perforation documented by radiological imaging during a therapeutic exam. Nine of 12 patients requiring an ER/ physician visit had esophageal pathology. Esophageal nar-
rowing (stricture or mass) was the finding associated with the greatest number of ER/physician visits (four patients) and hospitalizations (three patients). Dysphagia as an indication resulted in the most ER/physician visits (five patients) and hospitalizations (three patients). Specific Complications OROPHARYNGEAL PROBLEMS. Of our cohort, 10.1% (48 patients) reported oropharyngeal problems 30 days after their outpatient upper endoscopy that they felt was procedure-related. When oropharyngeal problems occurred, they always occurred within 24 h of the procedure. Of these patients, 94% (45 of 48) complained of a sore throat. The other three complained of a sore lip, hoarseness, and a torn uvula, respectively. Four of these patients, including the patient with a torn uvula after dilation, required an ER/ physician visit after their procedure but no patient required hospitalization. Of the women, 11% complained of a sore throat after their procedure, whereas 8% of men had this complaint. Of those patients complaining of sore throat the
Table 4. Procedures Performed in Patients With Complications Reported at 30 Days Procedure
Total Procedures Performed
30-Day Complications (%)
ER/Physician Visits (%)
Hospitalizations (%)
Confirmed Perforation
Diagnostic Therapeutic Dilation Thermal therapies Stent placement Injection Banding
370 97 60 32 6 4 2
64 (17.3%) 19 (19.6%) 7 (11.7%) 10 (31.3%) 1 (16.7%) 1 (25%) 2 (100%)
8 (2.16%) 4 (4.12%) 4 (6.7%) 0 1 (16.7%) 0 0
2 (0.54%) 3 (3.1%) 3 (5%) 0 1 (16.7%) 0 0
0 1 (1.0%) 1 (1.7%) 0 0 0 0
AJG – June, 1999
mean and median duration of this problem was 4.2 days and 2 days, respectively. No specific indication or therapy showed a significantly higher rate of oropharyngeal problems. Both patients who underwent variceal banding complained of sore throat, but the numbers of this subset were too small to draw conclusions. Of the patients with oropharyngeal problems, 64% had esophageal findings, whereas 46% of patients overall had esophageal findings. ABDOMINAL DISCOMFORT. Of our cohort, 5.3% (25 patients) reported abdominal discomfort 30 days after their outpatient upper endoscopy that they felt was procedurerelated. The onset of abdominal discomfort was always within 24 h of the procedure when it occurred. Three of these patients required an ER/physician visit, and one required hospitalization. The hospitalized patient also required surgery. Seven percent of the women complained of abdominal discomfort and 3% of the men had this complaint. The mean and median duration of abdominal discomfort in patients with this complaint was 4.8 days and 2 days, respectively. The patient complaining of the longest duration of abdominal discomfort (17 days) was the patient who required surgery. Of the patients with abdominal discomfort, 56% (14 of 25) had their procedure done for the indication of dyspepsia, whereas only 33% of the total cohort (156 of 473) had their procedure performed for this indication (p 5 0.03, x2 test). Of the patients with abdominal discomfort, 36% (nine of 25) had a normal exam, whereas only 22% of the total cohort (106 of 473) had a normal exam (p 5 NS). Patients who underwent thermal therapies, especially laser or argon plasma coagulation, had an increased risk of developing abdominal discomfort. Twelve percent of patients with abdominal discomfort underwent laser or APC therapy; however, only 4.4% of patients had these therapies performed. Complications Discussed at Our Morbidity and Mortality Conferences Ten patients (2.1%) in our cohort had complications reported by either physicians or nurses, which were discussed at our morbidity and mortality conferences. Cardiopulmonary complications accounted for 50% of the complications discussed at morbidity and mortality (five of 10). Hypoxemia was the most common complication (three patients). Two patients required an ER/physician visit, and two patients required hospitalization. One of the hospitalized patients underwent surgery. Although cardiopulmonary complications were most frequent, none of these patients required an ER/physician visit or hospitalization. Both of the patients requiring hospitalization had problems related to the mechanical aspects of the procedure. One patient perforated after dilation and the other had a small bowel obstruction, presumably due to insufflation of air. Fifty percent of complications discussed at our morbidity and mortality conferences (five of 10) were reported by nurses, but not physicians: this included one complication that re-
Complications After Upper Endoscopy
1543
sulted in hospitalization. Four of six cardiopulmonary or allergic complications were reported only by nurses. In some cases it is possible that physicians encouraged nurses to document complications that they did not document themselves.
DISCUSSION Overall, 18.2% of patients reported problems that they felt were related to their procedure 30 days after outpatient upper endoscopy, whereas 2.1% of patients were discussed at our monthly morbidity and mortality conferences. The majority of complications reported by patients did not require further medical attention. However, only 16.7% of patients requiring an ER/physician visit and 40% of patients requiring hospitalization were discussed at our morbidity and mortality conferences. Emergency room and/or physician visits occurred for 2.2% of patients after diagnostic exams and for 4.1% of patients after therapeutic exams. Hospitalizations occurred for 0.5% of patients after diagnostic exams and for 3.1% of patients after therapeutic exams. There are a number of reasons why patients may have reported more complications than those discussed at morbidity and mortality conferences. First, it is probable that the majority of complications reported by patients were mild and that the physician was unaware that they occurred. The patient may not have felt that further medical attention was necessary. Patients requiring an ER/physician visit could also not have been brought to the attention of the endoscopist. The health care provider examining the patient as a result of the complication may not have informed the endoscopist. It is important for the endoscopist to be easily accessible to the patient and surrounding physicians, should follow-up be necessary. This may increase the endoscopist’s awareness of complications that occur and facilitate early intervention, if necessary. Second, it is possible that the endoscopist knew of the complication, thought it was a complication, but did not report it. Underreporting of complications does appear to occur (7, 8). One reason for this is probably that physicians do not enter late complications into procedure reports that have previously been created. To create accurate complication records it is essential that the endoscopist be vigilant about recording late complications. Finally, it is possible that physicians were aware of complications reported by patients, but did not believe that they were “true” complications. It is possible that some of the complications reported by patients were not actually caused by the procedure, but rather were part of the underlying process that initiated the endoscopy, i.e., irritable bowel syndrome, esophagitis, etc. Although this may be true of some patients, it is only by understanding complications as perceived by patients that we can hope to maximize our understanding of upper endoscopy and minimize the discomfort that procedures may cause. We also feel that objective grading of complications is necessary to identify
1544
Zubarik et al.
complications, which are more likely to use medical resources and to result in poorer patient outcome. The most common complications reported by patients in this study were oropharyngeal problems (10.1%) and abdominal discomfort (5.3%). The vast majority of patients reporting oropharyngeal problems had a sore throat. Sore throat resulting from upper endoscopy lasted for a median of 2 days. This complication is not surprising, given the nature of the procedure, and is most likely due to mechanical trauma caused by the endoscope. The use of throat anesthetic sprays in our endoscopic unit depends on the preference of the endoscopist. It is not certain whether the use of topical pharyngeal anesthetics or smaller caliber endoscopes could decrease rates of reported sore throat after upper endoscopy (9, 10). Abdominal discomfort was the second most frequent complication reported by patients. The median duration of this discomfort was 2 days. Two distinct groups appeared more likely to report abdominal discomfort. The first group was comprised of patients who had the procedure done for dyspepsia, especially if the endoscopic exam was normal. This may represent patients with functional dyspepsia. The cause of abdominal discomfort in this patient population is uncertain, and it is unclear whether endoscopy causes more perceived abdominal discomfort than these patients have at baseline. However, none of these patients required either an ER/physician visit or hospitalization. Drossman et al. (11) examined 793 patients and found that patients who were concerned about abdominal discomfort before endoscopy were more likely to experience this discomfort during an endoscopic exam. This was particularly true for women and for younger patients. The second group of patients reporting higher rates of abdominal discomfort were those who had thermal therapy performed. The reason for this is uncertain, but may be due to subsequent ulceration. Our data suggest that women are more likely than men to report complications at 30 days. It is not certain why this occurs. Anatomical differences in the oropharynx and the stomach may predispose women to sustain more trauma from the passage of an endoscope, and to tolerate gastric distention less well. It is known that the size and stature of a person produce variation in the size and shape of the stomach (12). Another explanation, however, could be societal differences in the perception of discomfort and in the willingness to report it to an interviewer. These reasons are not mutually exclusive, and both could contribute to the disparity of reported complications between genders. There are some limitations to our study. First, the cohort design of the study allows us to show associations between such factors as endoscopic dilation and subsequent higher rates of hospitalization, but it does not allow us to demonstrate cause and effect. Second, it is not always certain whether the reported complications were due to the procedure performed or to the underlying process for which the patient was endoscoped. This criticism is valid, but we feel it is better to assume that the complications are related to the
AJG – Vol. 94, No. 6, 1999
procedure than to dismiss potential complications and to risk underestimating the effect that endoscopy has on our patients. If we had not included complications reported by patients at 30 days, the majority of complications resulting in an ER/physician visit or hospitalization would not have been detected for subsequent review. Third, our definition of a complication is not necessarily standard. We included complications as perceived by physicians, nurses, or patients. Ultimately, the goal is to minimize the detrimental effect of upper endoscopy while maximizing the benefit. Physicians, nurses, and patients view procedures from somewhat different perspectives, and subsequently report different complications. It is important for the physician to be aware of the different viewpoints of those involved. Fourth, we used ER/physician visits, hospitalizations and procedure-related mortality to grade the severity of complications. There are other factors, such as medical therapy required and disability resulting from a procedure, that are important outcomes. The O (outcome), D (disability), D (death) score, a recently developed grading system of endoscopic complications, addresses some of these other outcomes (13). This system enables the endoscopist to objectively grade complications that occur through a numeric system with three separate categories (outcome, disability, and death). The use of objective grading systems for endoscopic complications, such as the ODD score, should facilitate comparison of complications between centers and procedures. Finally, this study includes only the experience of one endoscopic center. Regional variation in the performance of procedures and patient populations may limit the generalizability of our data to other centers. Nationally used endoscopic databases, such as the clinical outcomes research initiative (CORI) database (14), could help to determine variation in endoscopic use and complication rates.
CONCLUSIONS Significantly more complications are detected by contacting patients 30 days after their outpatient upper endoscopy than are discussed at our morbidity and mortality conferences. The majority of complications that require an ER/physician visit or hospitalization are reported by patients at 30 days but are not discussed at our monthly morbidity and mortality conferences. ER/physician visits and hospitalizations were more frequent after therapeutic than after diagnostic exams. The most common complications reported by patients at 30 days are oropharyngeal problems (10.1%) and abdominal discomfort (5.3%). These complications should not alter use of upper endoscopy, but ways should be sought to minimize their occurrence. Women appear to report more total complications than men. The majority of patients requiring hospitalization had esophageal narrowing (stricture or mass) and required dilation. Multicenter efforts are needed to determine whether our data can be verified and generalized to other centers.
AJG – June, 1999
Complications After Upper Endoscopy
ACKNOWLEDGMENTS This study was funded by a grant from the American College of Gastroenterology. This work was presented in part as an oral presentation at the American College of Gastroenterology meeting in Chicago (1997), and as a poster at the Digestive Disease Week in New Orleans (1998). Reprint requests and correspondence: Richard Zubarik, M.D., 8987 Shady Leaf Cove, Cordova, TN 38018. Received Sep. 11, 1998; accepted Jan. 26, 1999.
REFERENCES 1. Silvis SE, Nebel O, Rogers G, et al. Endoscopic complications. Results of the 1974 American Society for Gastrointestinal Endoscopy Survey. JAMA 1976;235:928 –30. 2. Davis RE, Graham DY. Endoscopic complications. The Texas experience. Gastrointest Endosc 1979;25:146. 3. Fleischer DE, Al-Kawas F, Benjamin SB, et al. Prospective evaluation of complications in an endoscopy unit: Use of the A/S/G/E quality care guidelines. Gastrointest Endosc 1992; 38:411– 4. 4. Fleischer DE. Better definition of endoscopic complications and other negative outcomes. Gastrointest Endosc 1994;40: 511– 4. 5. Winslow CM, Solomon DH, Chassin MR, et al. The appro-
6. 7.
8. 9. 10. 11. 12. 13. 14.
1545
priateness of carotid endarterectomy. N Engl J Med 1988;318: 721–7. Newcomer MK, Jowell PS, Cotton PB, et al. Underestimation of adverse events following ERCP: A prospective 30 day follow-up study. Gastrointest Endosc 1995;41:408. Arrowsmith JB, Gerstman BB, Fleischer DE, et al. Results from the American Society for Gastrointestinal Endoscopy/ U.S. Food and Drug Administration collaborative study on complication rates and drug use during gastrointestinal endoscopy. Gastrointest Endosc 1991;37:421–7. Baillie J. Complications of endoscopy. Endoscopy 1994;26: 185–203. Canter DS, Baldridge ET. Premedication with meperidine and diazepam for gastrointestinal endoscopy precludes the need for topical anesthesia. Gastrointest Endosc 1986;32:339 – 41. Benjamin SB. Complications of conscious sedation. Gastrointest Endosc Clin North Am 1996;6:277– 86. Drossman DA, Brandt LJ, Sears C, et al. A preliminary study of patients’ concerns related to GI endoscopy. Am J Gastroenterol 1996;91:287–91. Woodburne RT, Burkel WE. Essentials of human anatomy, 9th ed. Oxford: University Press, 1994. Fleischer DE, Van de Mierop F, Eisen GM, et al. A new system for defining endoscopic complications emphasizing the measure of importance. Gastrointest Endosc 1997;45:128 –33. Lieberman DA, Fleischer DE, Eisen GM, et al. Clinical outcomes research initiative: Preliminary report of a clinical practice consortium. Gastrointest Endosc 1997;45:AB52 (abstract).
Vol. 94, No. 6, 1999 ISSN 0002-9270/99/$20.00 PII S0002-9270(99)00197-5
Prospective Analysis of Complications 30 Days After Outpatient Upper Endoscopy Richard Zubarik, M.D., Glenn Eisen, M.D., M.P.H., Chris Mastropietro, Jennifer Lopez, John Carroll, M.D., Stanley Benjamin, M.D., and David E. Fleischer, M.D. Department of Gastroenterology, Georgetown University Hospital, Washington, DC, and Gastroenterology Section, VA Medical Center, Memphis, Tennessee
OBJECTIVE: The aim of this study was to compare complication rates reported by patients 30 days after outpatient upper endoscopy with those discussed at our monthly morbidity and mortality conference. We also intended to establish which complications were reported most frequently 30 days after upper endoscopy, and which patients or procedures involved the highest risk. METHODS: Trained interviewers performed standardized telephone interviews on consecutive outpatients undergoing upper endoscopy over a 1-yr period. Patients were queried regarding potential events related to their upper endoscopy in the 30 days subsequent, including symptoms, emergency room (ER) and/or physician visits, and hospitalizations. The indications, findings, and therapies were reviewed from endoscopic reports. RESULTS: A total of 473 patients were contacted 30 days after outpatient upper endoscopy and agreed to participate in our study. The most common complications reported by patients at 30 days were sore throat (9.5%) and abdominal discomfort (5.3%). Twelve patients (2.5%) required an ER/ physician visit and five patients (1.1%) required hospitalization. The minority of both ER/physician visits (16.7%) and hospitalizations (40%) were discussed at our monthly morbidity and mortality conferences. CONCLUSIONS: More complications were reported by patients 30 days after outpatient upper endoscopy than were discussed at our monthly morbidity and mortality conferences. The most frequent complications reported by patients were sore throat and abdominal pain. The minority of ER/ physician visits and hospitalizations were discussed at our morbidity and mortality conferences. (Am J Gastroenterol 1999;94:1539 –1545. © 1999 by Am. Coll. of Gastroenterology)
members regarding endoscopic complications (1). The rate of complications after esophagogastroduodenoscopy (EGD) was 1.3 complications/1000 procedures. Data from this study were derived retrospectively and relied on physician recall or procedural reports. Only 64% of those surveyed returned questionnaires. A similar survey study was performed by the Texas Society for Gastrointestinal Endoscopy in 1979 (2). In this study, the rate of complications after EGD was 2.4 complications/1000 procedures. These studies are important, but they have limitations and may not accurately reflect complications that occur. A prospective study performed at our institution in 1990 evaluated complications discussed at morbidity and mortality conferences over a 1-yr period (3). The complication rate for endoscopic procedures performed was 1.9%. This study included all endoscopic procedures. Patients also were not systematically contacted after their procedure, so all complications may not have been identified. Ambiguity regarding the definition and gradation of complications is also a significant problem (4). Terms such as “major” and “minor” mean different things to different people, thereby making it difficult to compare complications from one study to the next. We used a 30-day follow-up period for procedure-related complications because it is similar to that used for surgical mortality rates (5) and because it has been used in the past to evaluate ERCPrelated complications (6). Complications were graded by recording whether they required an emergency room (ER) and/or physician visit, required hospitalization, or resulted in mortality. The primary question that this study addressed was whether complications were more frequently reported by patients 30 days after outpatient upper endoscopy than were discussed at our monthly morbidity and mortality conferences. We also attempted to determine the most common complications that patients reported at 30 days, and to see if any factors predicted higher risk.
INTRODUCTION There are surprisingly few studies analyzing the complications of upper endoscopy. In 1974, the American Society for Gastrointestinal Endoscopy performed a survey study of its This work was performed at Georgetown University Hospital, Washington, DC.
MATERIALS AND METHODS The Georgetown Upper Endoscopy Cohort (GUEC) consisted of patients undergoing outpatient upper endoscopy at Georgetown University Hospital from January 14, 1997, to January 13, 1998. Sixteen gastroenterologists had patients
1540
Zubarik et al.
AJG – Vol. 94, No. 6, 1999
Table 1. Complications at 30-Day Follow-Up Compared With Those Discussed at Morbidity and Mortality Conference
30-Day follow-up Morbidity and mortality 30-Day follow-up or morbidity and mortality
Procedures
Total Complications (%)
ER/Physician Visits (%)
Hospitalizations (%)
473 473 473
86 (18.2) 10 (2.1) 91 (19.2)
12 (2.5) 2 (0.4) 12 (2.5)
5 (1.1) 2 (0.4) 5 (1.1)
enrolled in our study. Nine were adult academic gastroenterologists, five were private gastroenterologists, and two were pediatric gastroenterologists. The majority of procedures were performed by adult academic gastroenterologists with or without the assistance of a gastroenterology fellow. Three trained interviewers performed standardized telephone interviews on consecutive outpatients undergoing upper endoscopy. Patients were contacted by telephone 30 days after their procedure. If a patient was not contacted on the initial attempt, six repeat attempts were made before excluding the patient from our study. The specific questions that patients were asked were: 1) Did you have any problems that you felt were related to the procedure? If yes, how long did this problem persist? 2) Have you gone to see a doctor or to an emergency room for this problem? 3) Have you been hospitalized for this problem? If yes, why? Patients were excluded if they underwent concomitant colonoscopy, flexible sigmoidoscopy, gastrostomy tube placement, enteroscopy, endoscopic ultrasound, or endoscopic retrograde cholangiopancreatography. Patients were also excluded if they could not speak English or if they did not have a telephone. The indications, findings, and therapies performed during upper endoscopy were reviewed from the endoscopic reports of those patients included in our study. Charts were reviewed of those patients requiring an ER/physician visit or hospitalization. Patients that required hospitalization were discussed with the attending physician who performed the involved procedure. Scheduled hospitalizations (i.e., after esophageal stent placement), or hospitalizations resulting from findings during a procedure (but not a complication) were excluded. Morbidity and Mortality Conferences Both physicians and nurses record endoscopic complications at Georgetown University. Physicians enter complications into their computer-generated endoscopic reports. This is done at the time of the procedure if the complication is immediate or later if it is delayed. Endoscopic reports are created either by the attending physician performing the procedure or by a gastroenterology fellow if one is involved. Nurses enter complications into an endoscopic log book in which every procedure is recorded. The complications reported by either physicians or nurses are then discussed at our monthly morbidity and mortality conferences at which attending physicians, fellows, residents, and nurses are present. All complications that are recorded either in the computerized reports or in the unit log book are discussed at our monthly morbidity and mortality conferences. Compli-
cations occurring during the procedure or afterward are discussed. Discussion of these complications occurs even if the involved attending physician does not come to the meeting. Definition A complication was defined as any deviation from the optimal course after upper endoscopy associated with the procedure as reported by a physician, nurse, or patient. We stratified the severity of reported complications by determining whether they required an ER/physician visit, required a hospitalization, or resulted in mortality.
RESULTS Of 627 patients undergoing outpatient upper endoscopy, 473 (75.4%) were successfully contacted and agreed to participate in our study. The average age of participating patients was 52.7 yr and 54.6% were female. The most common reason for exclusion from our study was inability to contact the patient by telephone; this accounted for 71% of patients who were excluded. This group included patients who did not answer (53 patients), who gave a wrong telephone number (20 patients), whose telephone number was disconnected (18 patients), who were out of town (13 patients), and who did not give a telephone number at all (6 patients). Seven patients who were excluded died before 30-day follow-up. None of these patients died as a result of endoscopy. Ninety-one patients had a complication that was either detected at 30-day follow-up or discussed at our morbidity and mortality conferences (Table 1). Eighty-six patients (18.2%) reported procedure-related problems at 30-day follow-up, and 10 patients (2.1%) were discussed at our morbidity and mortality conferences. Twelve patients required an ER/physician visit as a result of their procedure (2.5%), and five patients required hospitalization (1.1%). There was no procedure-related mortality. Only 16.7% (two of 12) of the patients requiring an ER/physician visit, and 40% (two of five) of the patients requiring hospitalization detected at 30-day follow-up were discussed at morbidity and mortality conferences. Three of the five hospitalized patients had undergone dilation, and were hospitalized to rule out perforation. Two of these patients had fever and one had chest pain (Table 2). One of the hospitalized patients who had undergone dilation had migration of an esophageal stent and subsequently required repeat dilation, stent removal, and placement of another stent. Of the other two patients hospitalized after esophageal
AJG – June, 1999
Complications After Upper Endoscopy
1541
Table 2. Patients Who Required an ER/Physician Visit or Hospitalization After Outpatient Upper Endoscopy Known by Endoscopist
ER/Physician Visit
Hospitalization
Surgery
chest pain
yes
yes
yes
no
yes
fever
yes
yes
yes
no
yes
yes
yes
yes
yes
no
abdominal distention fever
yes
yes
yes
no
none dilation
no no
presyncope torn uvula
no yes
yes yes
yes no
no no
none
no
sore throat
no
yes
no
no
none none
no no
sore throat sore throat
no no
yes yes
no no
no no
none
no
no
yes
no
no
none
no
no
yes
no
no
none
no
abdominal pain abdominal pain chest pain
no
yes
no
no
Patient
Finding
Therapy
1
migrated stent esophageal stricture esophagitis
dilation/stent replacement dilation none
anastomotic stricture gastritis esophageal stricture radiation esophagitis normal cicatricial pemphigoid gastritis
dilation
gastric polyps barrett mucosa
2 3 4 5 6 7 8 9 10 11 12
Reported at M&M no
30-Day Complication
M&M 5 morbidity and mortality conference.
dilation, one had balloon dilation and the other had dilation of a benign stricture. Only one of the three had a documented perforation on radiological imaging (the benign stricture). In this patient the perforation was confined, and did not require surgery. Two patients hospitalized after outpatient upper endoscopy had not undergone esophageal dilation. One of these patients was admitted for presyncope and the other was admitted for a small bowel obstruction. The small bowel obstruction began directly after the procedure in a patient with AIDS, and required surgical intervention the next day after worsening symptoms. The initial upper endoscopy revealed only esophagitis and, at the time of surgery, only adhesions were found.
Thirty-Day Follow-Up A total of 86 patients (18.2%) complained of 95 complications 30 days after outpatient upper endoscopy (Table 3). Oropharyngeal problems (48 patients) and abdominal discomfort (22 patients) were the most common complications reported. Oropharyngeal problems accounted for the most ER/physician visits (four patients), and fever resulted in the most hospitalizations (two patients). Women were more likely to report complications at 30 days than were men (22.5% vs 13%, p , 0.001, x2 test). There was no significant gender difference in ER/physician visits or hospitalizations (Fig. 1). No specific indications, findings, or therapies predicted a significantly higher rate of total complications at
Table 3. Complications 30 Days After Outpatient Upper Endoscopy
Complication
Complications at 30-Day Follow-Up
ER/Physician Visits
Hospitalizations
Oropharyngeal complications (total) Sore throat Sore lip Torn uvula Hoarse Abdominal discomfort (total) Abdominal pain Abdominal distention Nausea/Vomiting Chest Pain Fever/Chills Fatigue Presyncope Other Total
48 45 1 1 1 25 21 4 7 3 3 2 2 5 95
4 3 0 1 0 3 2 1 0 2 2 0 1 0 12
0 0 0 0 0 1 0 1 0 1 2 0 1 0 5
1542
Zubarik et al.
AJG – Vol. 94, No. 6, 1999
Figure 1. Complications by gender.
thirty days. Patients having upper endoscopy performed for the indication of dyspepsia, who had a normal exam, had a higher rate of total reported complications at 30 days (25.5%). This difference was not statistically significant, however, and none of these patients required an ER/physician visit or hospitalization. Reported total complication rates at 30 days were slightly higher for therapeutic examinations than for diagnostic examinations (19.6% vs 17.3%). This was also true for ER/physician visits and hospitalizations. When diagnostic procedures were performed, 2.2% of patients required an ER/physician visit and 0.5% of patients required hospitalization (Table 4). When therapeutic exams were performed, 4.1% of patients required an ER/physician visit, and 3.1% required hospitalization. The therapy resulting in the most hospitalizations was endoscopic dilation, which had an overall hospitalization rate of 5%. None of the diagnostic exams resulted in perforation. There was one perforation documented by radiological imaging during a therapeutic exam. Nine of 12 patients requiring an ER/ physician visit had esophageal pathology. Esophageal nar-
rowing (stricture or mass) was the finding associated with the greatest number of ER/physician visits (four patients) and hospitalizations (three patients). Dysphagia as an indication resulted in the most ER/physician visits (five patients) and hospitalizations (three patients). Specific Complications OROPHARYNGEAL PROBLEMS. Of our cohort, 10.1% (48 patients) reported oropharyngeal problems 30 days after their outpatient upper endoscopy that they felt was procedure-related. When oropharyngeal problems occurred, they always occurred within 24 h of the procedure. Of these patients, 94% (45 of 48) complained of a sore throat. The other three complained of a sore lip, hoarseness, and a torn uvula, respectively. Four of these patients, including the patient with a torn uvula after dilation, required an ER/ physician visit after their procedure but no patient required hospitalization. Of the women, 11% complained of a sore throat after their procedure, whereas 8% of men had this complaint. Of those patients complaining of sore throat the
Table 4. Procedures Performed in Patients With Complications Reported at 30 Days Procedure
Total Procedures Performed
30-Day Complications (%)
ER/Physician Visits (%)
Hospitalizations (%)
Confirmed Perforation
Diagnostic Therapeutic Dilation Thermal therapies Stent placement Injection Banding
370 97 60 32 6 4 2
64 (17.3%) 19 (19.6%) 7 (11.7%) 10 (31.3%) 1 (16.7%) 1 (25%) 2 (100%)
8 (2.16%) 4 (4.12%) 4 (6.7%) 0 1 (16.7%) 0 0
2 (0.54%) 3 (3.1%) 3 (5%) 0 1 (16.7%) 0 0
0 1 (1.0%) 1 (1.7%) 0 0 0 0
AJG – June, 1999
mean and median duration of this problem was 4.2 days and 2 days, respectively. No specific indication or therapy showed a significantly higher rate of oropharyngeal problems. Both patients who underwent variceal banding complained of sore throat, but the numbers of this subset were too small to draw conclusions. Of the patients with oropharyngeal problems, 64% had esophageal findings, whereas 46% of patients overall had esophageal findings. ABDOMINAL DISCOMFORT. Of our cohort, 5.3% (25 patients) reported abdominal discomfort 30 days after their outpatient upper endoscopy that they felt was procedurerelated. The onset of abdominal discomfort was always within 24 h of the procedure when it occurred. Three of these patients required an ER/physician visit, and one required hospitalization. The hospitalized patient also required surgery. Seven percent of the women complained of abdominal discomfort and 3% of the men had this complaint. The mean and median duration of abdominal discomfort in patients with this complaint was 4.8 days and 2 days, respectively. The patient complaining of the longest duration of abdominal discomfort (17 days) was the patient who required surgery. Of the patients with abdominal discomfort, 56% (14 of 25) had their procedure done for the indication of dyspepsia, whereas only 33% of the total cohort (156 of 473) had their procedure performed for this indication (p 5 0.03, x2 test). Of the patients with abdominal discomfort, 36% (nine of 25) had a normal exam, whereas only 22% of the total cohort (106 of 473) had a normal exam (p 5 NS). Patients who underwent thermal therapies, especially laser or argon plasma coagulation, had an increased risk of developing abdominal discomfort. Twelve percent of patients with abdominal discomfort underwent laser or APC therapy; however, only 4.4% of patients had these therapies performed. Complications Discussed at Our Morbidity and Mortality Conferences Ten patients (2.1%) in our cohort had complications reported by either physicians or nurses, which were discussed at our morbidity and mortality conferences. Cardiopulmonary complications accounted for 50% of the complications discussed at morbidity and mortality (five of 10). Hypoxemia was the most common complication (three patients). Two patients required an ER/physician visit, and two patients required hospitalization. One of the hospitalized patients underwent surgery. Although cardiopulmonary complications were most frequent, none of these patients required an ER/physician visit or hospitalization. Both of the patients requiring hospitalization had problems related to the mechanical aspects of the procedure. One patient perforated after dilation and the other had a small bowel obstruction, presumably due to insufflation of air. Fifty percent of complications discussed at our morbidity and mortality conferences (five of 10) were reported by nurses, but not physicians: this included one complication that re-
Complications After Upper Endoscopy
1543
sulted in hospitalization. Four of six cardiopulmonary or allergic complications were reported only by nurses. In some cases it is possible that physicians encouraged nurses to document complications that they did not document themselves.
DISCUSSION Overall, 18.2% of patients reported problems that they felt were related to their procedure 30 days after outpatient upper endoscopy, whereas 2.1% of patients were discussed at our monthly morbidity and mortality conferences. The majority of complications reported by patients did not require further medical attention. However, only 16.7% of patients requiring an ER/physician visit and 40% of patients requiring hospitalization were discussed at our morbidity and mortality conferences. Emergency room and/or physician visits occurred for 2.2% of patients after diagnostic exams and for 4.1% of patients after therapeutic exams. Hospitalizations occurred for 0.5% of patients after diagnostic exams and for 3.1% of patients after therapeutic exams. There are a number of reasons why patients may have reported more complications than those discussed at morbidity and mortality conferences. First, it is probable that the majority of complications reported by patients were mild and that the physician was unaware that they occurred. The patient may not have felt that further medical attention was necessary. Patients requiring an ER/physician visit could also not have been brought to the attention of the endoscopist. The health care provider examining the patient as a result of the complication may not have informed the endoscopist. It is important for the endoscopist to be easily accessible to the patient and surrounding physicians, should follow-up be necessary. This may increase the endoscopist’s awareness of complications that occur and facilitate early intervention, if necessary. Second, it is possible that the endoscopist knew of the complication, thought it was a complication, but did not report it. Underreporting of complications does appear to occur (7, 8). One reason for this is probably that physicians do not enter late complications into procedure reports that have previously been created. To create accurate complication records it is essential that the endoscopist be vigilant about recording late complications. Finally, it is possible that physicians were aware of complications reported by patients, but did not believe that they were “true” complications. It is possible that some of the complications reported by patients were not actually caused by the procedure, but rather were part of the underlying process that initiated the endoscopy, i.e., irritable bowel syndrome, esophagitis, etc. Although this may be true of some patients, it is only by understanding complications as perceived by patients that we can hope to maximize our understanding of upper endoscopy and minimize the discomfort that procedures may cause. We also feel that objective grading of complications is necessary to identify
1544
Zubarik et al.
complications, which are more likely to use medical resources and to result in poorer patient outcome. The most common complications reported by patients in this study were oropharyngeal problems (10.1%) and abdominal discomfort (5.3%). The vast majority of patients reporting oropharyngeal problems had a sore throat. Sore throat resulting from upper endoscopy lasted for a median of 2 days. This complication is not surprising, given the nature of the procedure, and is most likely due to mechanical trauma caused by the endoscope. The use of throat anesthetic sprays in our endoscopic unit depends on the preference of the endoscopist. It is not certain whether the use of topical pharyngeal anesthetics or smaller caliber endoscopes could decrease rates of reported sore throat after upper endoscopy (9, 10). Abdominal discomfort was the second most frequent complication reported by patients. The median duration of this discomfort was 2 days. Two distinct groups appeared more likely to report abdominal discomfort. The first group was comprised of patients who had the procedure done for dyspepsia, especially if the endoscopic exam was normal. This may represent patients with functional dyspepsia. The cause of abdominal discomfort in this patient population is uncertain, and it is unclear whether endoscopy causes more perceived abdominal discomfort than these patients have at baseline. However, none of these patients required either an ER/physician visit or hospitalization. Drossman et al. (11) examined 793 patients and found that patients who were concerned about abdominal discomfort before endoscopy were more likely to experience this discomfort during an endoscopic exam. This was particularly true for women and for younger patients. The second group of patients reporting higher rates of abdominal discomfort were those who had thermal therapy performed. The reason for this is uncertain, but may be due to subsequent ulceration. Our data suggest that women are more likely than men to report complications at 30 days. It is not certain why this occurs. Anatomical differences in the oropharynx and the stomach may predispose women to sustain more trauma from the passage of an endoscope, and to tolerate gastric distention less well. It is known that the size and stature of a person produce variation in the size and shape of the stomach (12). Another explanation, however, could be societal differences in the perception of discomfort and in the willingness to report it to an interviewer. These reasons are not mutually exclusive, and both could contribute to the disparity of reported complications between genders. There are some limitations to our study. First, the cohort design of the study allows us to show associations between such factors as endoscopic dilation and subsequent higher rates of hospitalization, but it does not allow us to demonstrate cause and effect. Second, it is not always certain whether the reported complications were due to the procedure performed or to the underlying process for which the patient was endoscoped. This criticism is valid, but we feel it is better to assume that the complications are related to the
AJG – Vol. 94, No. 6, 1999
procedure than to dismiss potential complications and to risk underestimating the effect that endoscopy has on our patients. If we had not included complications reported by patients at 30 days, the majority of complications resulting in an ER/physician visit or hospitalization would not have been detected for subsequent review. Third, our definition of a complication is not necessarily standard. We included complications as perceived by physicians, nurses, or patients. Ultimately, the goal is to minimize the detrimental effect of upper endoscopy while maximizing the benefit. Physicians, nurses, and patients view procedures from somewhat different perspectives, and subsequently report different complications. It is important for the physician to be aware of the different viewpoints of those involved. Fourth, we used ER/physician visits, hospitalizations and procedure-related mortality to grade the severity of complications. There are other factors, such as medical therapy required and disability resulting from a procedure, that are important outcomes. The O (outcome), D (disability), D (death) score, a recently developed grading system of endoscopic complications, addresses some of these other outcomes (13). This system enables the endoscopist to objectively grade complications that occur through a numeric system with three separate categories (outcome, disability, and death). The use of objective grading systems for endoscopic complications, such as the ODD score, should facilitate comparison of complications between centers and procedures. Finally, this study includes only the experience of one endoscopic center. Regional variation in the performance of procedures and patient populations may limit the generalizability of our data to other centers. Nationally used endoscopic databases, such as the clinical outcomes research initiative (CORI) database (14), could help to determine variation in endoscopic use and complication rates.
CONCLUSIONS Significantly more complications are detected by contacting patients 30 days after their outpatient upper endoscopy than are discussed at our morbidity and mortality conferences. The majority of complications that require an ER/physician visit or hospitalization are reported by patients at 30 days but are not discussed at our monthly morbidity and mortality conferences. ER/physician visits and hospitalizations were more frequent after therapeutic than after diagnostic exams. The most common complications reported by patients at 30 days are oropharyngeal problems (10.1%) and abdominal discomfort (5.3%). These complications should not alter use of upper endoscopy, but ways should be sought to minimize their occurrence. Women appear to report more total complications than men. The majority of patients requiring hospitalization had esophageal narrowing (stricture or mass) and required dilation. Multicenter efforts are needed to determine whether our data can be verified and generalized to other centers.
AJG – June, 1999
Complications After Upper Endoscopy
ACKNOWLEDGMENTS This study was funded by a grant from the American College of Gastroenterology. This work was presented in part as an oral presentation at the American College of Gastroenterology meeting in Chicago (1997), and as a poster at the Digestive Disease Week in New Orleans (1998). Reprint requests and correspondence: Richard Zubarik, M.D., 8987 Shady Leaf Cove, Cordova, TN 38018. Received Sep. 11, 1998; accepted Jan. 26, 1999.
REFERENCES 1. Silvis SE, Nebel O, Rogers G, et al. Endoscopic complications. Results of the 1974 American Society for Gastrointestinal Endoscopy Survey. JAMA 1976;235:928 –30. 2. Davis RE, Graham DY. Endoscopic complications. The Texas experience. Gastrointest Endosc 1979;25:146. 3. Fleischer DE, Al-Kawas F, Benjamin SB, et al. Prospective evaluation of complications in an endoscopy unit: Use of the A/S/G/E quality care guidelines. Gastrointest Endosc 1992; 38:411– 4. 4. Fleischer DE. Better definition of endoscopic complications and other negative outcomes. Gastrointest Endosc 1994;40: 511– 4. 5. Winslow CM, Solomon DH, Chassin MR, et al. The appro-
6. 7.
8. 9. 10. 11. 12. 13. 14.
1545
priateness of carotid endarterectomy. N Engl J Med 1988;318: 721–7. Newcomer MK, Jowell PS, Cotton PB, et al. Underestimation of adverse events following ERCP: A prospective 30 day follow-up study. Gastrointest Endosc 1995;41:408. Arrowsmith JB, Gerstman BB, Fleischer DE, et al. Results from the American Society for Gastrointestinal Endoscopy/ U.S. Food and Drug Administration collaborative study on complication rates and drug use during gastrointestinal endoscopy. Gastrointest Endosc 1991;37:421–7. Baillie J. Complications of endoscopy. Endoscopy 1994;26: 185–203. Canter DS, Baldridge ET. Premedication with meperidine and diazepam for gastrointestinal endoscopy precludes the need for topical anesthesia. Gastrointest Endosc 1986;32:339 – 41. Benjamin SB. Complications of conscious sedation. Gastrointest Endosc Clin North Am 1996;6:277– 86. Drossman DA, Brandt LJ, Sears C, et al. A preliminary study of patients’ concerns related to GI endoscopy. Am J Gastroenterol 1996;91:287–91. Woodburne RT, Burkel WE. Essentials of human anatomy, 9th ed. Oxford: University Press, 1994. Fleischer DE, Van de Mierop F, Eisen GM, et al. A new system for defining endoscopic complications emphasizing the measure of importance. Gastrointest Endosc 1997;45:128 –33. Lieberman DA, Fleischer DE, Eisen GM, et al. Clinical outcomes research initiative: Preliminary report of a clinical practice consortium. Gastrointest Endosc 1997;45:AB52 (abstract).