- Email: [email protected]
A comparison of transnasal and transoral endoscopy with small-diameter endoscopes in unsedated patients
A comparison of transnasal and transoral endoscopy with small-diameter endoscopes in unsedated patients Alexander Craig, MBBS, Jane Hanlon, DipAppSc Nursing, John Dent, MRCP, FRACP, PhD, Mark Schoeman, MBBS, FRACP, PhD Adelaide, Australia
Background: The aim of this study was to compare use and tolerance of transnasal and transoral diagnostic endoscopy with small-diameter endoscopes in unsedated patients. Methods: Patients being seen for diagnostic endoscopy were randomly assigned to have an unsedated transnasal or transoral procedure. Two prototype narrowdiameter endoscopes (a 5.3 mm fiberoptic endoscope and a 5.9 mm videoendoscope) were used. Results: Of 170 patients (64 women and 106 men) enrolled, 86 underwent transoral and 84 underwent transnasal endoscopy. The procedure was successfully completed using the transoral route in 85 of 86 patients and using the transnasal route in 74 of 84 patients (p = 0.004). Sixteen patients experienced mild epistaxis after transnasal endoscopy.The larger videoendoscope, when compared with the fiberoptic endoscope, accounted for a significantly higher proportion of failures (8 of 41 vs. 2 of 43, p = 0.046) and cases of epistaxis (12 of 33 vs. 4 of 41, p = 0.007). The time taken for transoral endoscopy was shorter than the transnasal route (oral mean 13.7 ± 0.5 minutes, nasal mean 15.2 ± 0.6 minutes, p = 0.054). There was no difference between the 2 groups with respect to tolerance of the procedure. From the endoscopist’s perspective, the only statistically significant difference between the 2 groups was that endoscope insertion was easier by the oral route (p = 0.007). Conclusions: Unsedated transnasal endoscopy with the videoendoscope was less successful compared with the transnasal fiberoptic instrument or when compared with either instrument passed transorally. Use of the larger diameter videoendoscope also resulted in significantly more epistaxis in the transnasal endoscopy group. Endoscopists find transoral introduction of the endoscope easier; this may reflect their relative unfamiliarity with the nasal route. Once intubation has been successfully achieved for either route, patient tolerance is the same. (Gastrointest Endosc 1999;49:292-6.) Esophagogastroduodenoscopy (EGD) is a commonly performed and well-established diagnostic procedure. Standard EGD involves oral passage of a flexible endoscope, 9 mm in diameter, after the oropharynx has been anesthetized with a local anesReceived July 22, 1998. For revision August 20, 1998. Accepted October 21, 1998. From the Department of Gastrointestinal Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia. The data were presented at the Poster Session at the American Society for Gastrointestinal Endoscopy Annual Meeting during DDW in New Orleans May 20, 1998. Published in abstract form in Gastrointestinal Endoscopy 1998;47:AB 28. Reprint requests: Mark Schoeman, PhD, Department of Gastrointestinal Medicine, Royal Adelaide Hospital. North Tce, Adelaide, South Australia, Australia 5000. Copyright © 1999 by the American Society for Gastrointestinal Endoscopy 0016-5107/99/$8.00 + 0 37/1/95331 292
GASTROINTESTINAL ENDOSCOPY
thetic spray or gargle. In Australia, intravenous sedation is given routinely to reduce patient discomfort and anxiety. However, sedation can lead to respiratory depression, and after the procedure patients require supervised recovery, which is sometimes prolonged. There are also significant indirect costs associated with the use of sedation because patients are unable to drive or to return to work for at least 12 hours after the procedure. Narrow-diameter endoscopes were developed in the early 1990s in the hope that the small caliber (< 6 mm) would reduce patient discomfort. The nasal route of insertion in particular was promoted because it was thought that this method would reduce pharyngeal discomfort and the risk of aspiration. Another potential advantage of nasal insertion is that patients are able to talk during the procedure, which can help reduce their anxiety. In addition, the VOLUME 49, NO. 3, PART 1, 1999
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
nasal route eliminates the risk that a patient will bite the endoscope. Initial studies in healthy volunteers 1 and in a cancer screening program in Japan 2 have shown that the transnasal unsedated procedure is safe and well tolerated, and that it allows adequate visualization of the upper gastrointestinal tract. Two studies directly compared unsedated transnasal EGD using a narrow diameter endoscope with standard sedated transoral EGD 3,4 and found acceptable patient tolerance and shortened recovery times with the first method. There is, however, some evidence from these earlier studies that the transnasal route is associated with an increase in adverse events such as epistaxis (13% in one study4), and it is our impression that some patients are unable to tolerate nasal intubation. In addition, endoscopists in our unit using the narrow-diameter endoscopes in unsedated patients have tended to choose the oral route of intubation because of its greater familiarity; patients are often unprepared to have a transnasal procedure. One study, published in abstract form only, directly compared the transnasal and the transoral routes of insertion for unsedated endoscopy with a narrowdiameter endoscope.5 This study, which included 50 patients, showed that the transoral route was better tolerated than the transnasal route. The aim of our study was to compare the transoral and the transnasal routes of insertion for unsedated EGD using narrow-diameter instruments in a large patient group, with respect to patient tolerance and endoscopists’ impressions. PATIENTS AND METHODS Entry into the trial was offered to all our patients aged 18 to 90 years referred for EGD to any of 8 endoscopists at our hospital from June 1996 to March 1998. After informed written consent was obtained, patients were randomly assigned to either transnasal or transoral endoscopy. We randomized the route of insertion using random numbers. The selected route was then sealed in an envelope, and the envelopes were sequentially numbered. After informed written consent was obtained, the next envelope was opened to determine the route of insertion. Patients were excluded from the study if there was a history of significant nasal trauma or surgery, if they were having a colonoscopy at the same appointment, if it was judged that a therapeutic procedure was likely to be required, or if there was a bleeding diathesis. Patients undergoing the transnasal procedure received a local anesthetic spray (lidocaine) to the nostril being intubated. Patients undergoing the transoral procedure received a local anesthetic gargle (amethocaine hydrochloride). Two prototype narrow-diameter endoscopes were used in the study (Olympus Australia, Oakleigh, Victoria): a 5.3 mm fiberoptic instrument (Fig. 1) and a 5.9 mm videoendoscope. Both endoscopes had a 2 mm accessory channel and VOLUME 49, NO. 3, PART 1, 1999
A Craig, J Hanlon, J Dent
Figure 1. Relative diameters of 3 endoscopes. Left, therapeutic endoscope, diameter 11 mm; middle, standard endoscope, diameter 9 mm; right, fiberoptic endoscope, 5.3 mm diameter. had similar performance characteristics. All endoscopists had prior experience with unsedated narrow-diameter endoscopy performed through either route. All procedures were performed with the patient in the left lateral position. Sedative medications were not routinely used, although atropine and prophylactic antibiotics could be used at the endoscopist’s discretion. An intravenous cannula was inserted only if atropine or antibiotics were given. All patients were monitored with pulse oximetry, and oxygen was administered if the oxygen saturation fell below 90%. The oxygen could be given via the nose or mouth depending on the route of endoscope insertion. After the procedure, the patient’s condition was monitored in a purpose-designed recovery room; fitness for discharge was assessed by using the modified recovery room American Pediatric Gross Assessment Record score. Data were collected from patients, the nursing staff and endoscopists. Specific data collected by the nursing staff included the time the patient entered and left the procedure room, the time spent in the recovery area before leaving, whether supplemental oxygen was required, whether any adverse event(s) occurred, and whether biopsies were performed. Patients were asked to score the procedure with visual analog scales. They scored how they tolerated the procedure, how well they generally felt, and how capable they felt about returning to work or normal activities. A visual analog scale was also used to assess the patient’s opinions on whether they would prefer an unsedated or sedated procedure if they required EGD in the future. Patients who had previously undergone a standard sedated procedure were asked whether they would want any future EGD performed with or without sedation. Endoscopist satisfaction was also assessed with visual analog scales for the level of difficulty encountered in intubation, for the adequacy of the view, for the performance of the endoscope with regard to air insufflation, the washing of the lens, and suction, for the degree of endoscope looping and for whether difficulty was encountered in entering the duodenum. If biopsy specimens were taken, the endoscopist was asked to rate the ease with which this task was performed. The study was reviewed and approved by the ethics committee of our hospital. Binary data were analyzed by means of chi-square tests of independence. Other data were analyzed by using GASTROINTESTINAL ENDOSCOPY
293
A Craig, J Hanlon, J Dent
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
Table 1. Procedure data Transoral
Transnasal
p Value
No. patients Age (years) Males Females Successful completion Procedure time (mean in minutes ± SE) Recovery time (mean in minutes ± SE) Adverse events
86 48.4 (19-84) 59 (69%) 27 (31%) 85 of 86 13.7 ± 0.5
84 51.1 (18-88) 47 (56%) 37 (44%) 74 of 84 15.2 ± 0.6
NS NS NS 0.004 0.054
10.0 ± 1.1
13.0 ± 1.4
NS
n=0
Supplemental O2 Biopsies
0 56
n=16 (epistaxis) 2 41
NS
Table 2. Mean patient response scores (out of 10 with 10 being the most positive response) ± standard error How did you tolerate the test? How do you feel? Could you return to work now? Would you be happy to have the test repeated in the form that you had it?
Transoral
Transnasal
p Value
6.4 ± 0.2
6.3 ± 0.3
NS
8.1 ± 0.3 8.6 ± 0.3
7.9 ± 0.3 8.4 ± 0.3
NS NS
9.0 ± 0.3
8.8 ± 0.3
NS
Patients who did not undergo a complete procedure were excluded. NS, Not significant
NS, Not significant
the Anova test. A p value less than 0.05 was taken as indicative of significant differences.
RESULTS The study included 170 patients (64 women, 106 men). The transoral route was used in 86 patients, and the transnasal route was used in 84 patients. The two groups were evenly matched for age (Table 1). The transoral group contained a higher proportion of men than did the transnasal group (Table 1), but this difference was not statistically significant. Similar numbers of patients underwent endoscopy with the fiberoptic endoscope and videoendoscope in each group (oral: video 42, fiberoptic 44; nasal: video 41, fiberoptic 43). The procedure was successfully completed in 85 of 86 patients by the transoral route and in 74 of 84 patients by the transnasal route (p = 0.004). Failed intubations in the transnasal group were caused by a combination of difficult insertion of the endoscope through the nasal passage and patient intolerance. Failed intubations occurred at a steady rate throughout the study period and were equally distributed among endoscopists. Patients who could not complete the procedure were excluded from any subsequent analysis. No complications were noted in the patients who had the transoral route, whereas 16 of 74 patients had epistaxis after the transnasal route. In 1 patient, the epistaxis continued for 8 hours but did not necessitate any specific intervention. In all other patients the epistaxis stopped within 5 minutes. No other adverse events were recorded. Two patients undergoing the transnasal route required supplemental oxygen for transient oxygen desaturation. The time taken for the endoscopy and the time spent in the recovery room are shown in Table 1. 294
GASTROINTESTINAL ENDOSCOPY
There was a trend toward a shorter procedure time for the transoral route. There was no difference between the two groups in the time spent in the recovery room. Patients in the transnasal group with epistaxis had a recovery time similar to that of the other patients in this group (mean 13.0 minutes). Patient assessments of the procedure are summarized in Table 2. There was no difference between the two groups with respect to any of the factors evaluated. Fifty-four patients had previously undergone a standard transoral sedated endoscopy, and 49 of these patients (oral: 25, nasal: 24) said that they would prefer to undergo an unsedated endoscopy with a narrow-diameter endoscope if required in the future. Endoscopist assessments are shown in Table 3. From the endoscopists’ perspective, the only statistically significant difference between the two groups was that endoscope insertion was judged to be easier by the oral route (p = 0.007). The difference between assessments of the adequacy of the endoscopic view almost reached statistical significance (p = 0.0501), with transoral being better than transnasal. With respect to ability to insufflate air, wash the lens, suction, instrument looping, and the ability to enter the duodenum there were no differences between the two groups. Biopsies were performed in 97 patients with no difference in the endoscopist score for performing biopsies (p = 0.7). All biopsies performed were adequate for histopathologic assessment. A subanalysis was performed on the video and fiberoptic instruments for the transnasal route and is summarized in Table 4. This disclosed a significantly lower rate of successful completion of the procedure with the videoendoscope (33 of 41 vs. 41 of VOLUME 49, NO. 3, PART 1, 1999
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
Table 3. Mean proceduralist scores (out of 10 with 10 being the most positive response) ± standard error Transoral Scope insertion View Insufflation Wash lens Suction Enter duodenum Looping
8.9 8.6 8.8 7.6 8.5 9.9 8.3
± ± ± ± ± ± ±
0.3 0.3 0.3 0.4 0.3 1.3 0.3
Transnasal 8.1 7.9 8.6 8.1 7.9 8.7 8.2
± ± ± ± ± ± ±
0.2 0.3 0.3 0.3 0.3 0.3 0.3
p value 0.007 0.0501 NS NS NS NS NS
43, p = 0.046) and a higher rate of epistaxis (12 of 33 vs. 4 of 41, p = 0.007). DISCUSSION This is the largest study that has specifically addressed the question of which route is best for insertion of a narrow-diameter endoscope in unsedated patients undergoing EGD. We found a higher rate of successful intubation with the transoral route. The transoral route was also the preferred route of insertion from the endoscopists’ perspective. One factor to account for these findings may be endoscopists’ relative unfamiliarity with the transnasal route, as the majority of routine endoscopic procedures are performed using the transoral route. Indeed, physicians who perform bronchoscopy would find a 10% failure rate for intubation with bronchoscopy unacceptable, although sedation is often used in this setting and bronchoscopes are still smaller than the narrow-diameter endoscopes. Once intubation had been successfully completed, patient tolerance was the same for both routes. The high rate of epistaxis (22.6%) may again be related to the endoscopists’ relative unfamiliarity with the transnasal route with resultant trauma to the nasal passage. Another explanation could be that the local anesthetic spray causes irritation. We did not use cocaine nasal spray, which is routinely used in otolaryngology and by chest physicians, as it has vasoconstrictive effects that probably reduce the incidence of bleeding. Epistaxis that is self-limited and does not necessitate intervention following transnasal bronchoscopy is reported as “common” but the incidence of significant epistaxis is low: 0.1% to 0.3%.6 All the episodes of epistaxis in this study resolved spontaneously. In the subanalysis of endoscope type, the failure of nasal intubation and the incidence of epistaxis was more common with the videoendoscope (5.9 mm) compared with the smaller fiberoptic (5.3 mm) VOLUME 49, NO. 3, PART 1, 1999
A Craig, J Hanlon, J Dent
Table 4. Sub-analysis of transnasal route between fiberoptic and video endoscopes Transnasal route Successful completion Epistaxis
Fiberoptic endoscope
Video endoscope
p Value
41 of 43 4 of 41
33 of 41 12 of 33
0.046 0.007
NS, Not significant.
endoscope. The fiberoptic instrument was passed with equal ease and with a comparable complication rate through either the mouth or the nose. This indicates that transnasal endoscopy should be performed only with endoscopes that are of a diameter of 5.3 mm or less. The larger videoendoscope, 5.9 mm in diameter, should be used only for unsedated transoral endoscopy. One advantage of unsedated transnasal endoscopy is a reduction in the risk of aspiration during the procedure compared with transoral endoscopy in sedated patients.1,3 During standard endoscopy, pharyngeal irritation may induce emesis which, when combined with oropharyngeal anesthesia, would predispose to aspiration. Transnasal endoscopy avoids these problems. In our study, there were no clinically evident aspiration events in either patient group. It is likely that a much larger study will be required to answer this question because the incidence of clinically evident and significant episodes of aspiration is low. It is our impression that patient sedation is the major contributor to occurrences of aspiration. Therefore, unsedated procedures are likely to be associated with a low risk of aspiration—patients are able to guard their airway more effectively. This study also allowed us to obtain further data that compare patient experiences of conventional sedated EGD with unsedated EGD performed with a narrow-diameter endoscope. We found that 90.7% of patients who had previously undergone sedated conventional EGD would prefer an unsedated EGD with a narrow-diameter endoscope in the future if required. Nozaki et al.2 reported a similar percentage of patients who found unsedated transnasal EGD with a small-diameter endoscope more comfortable than conventional sedated EGD. Our study also confirms that the performance characteristics of the narrow endoscopes are acceptable. The trend in this study toward better endoscopic views with the transoral route may be related to endoscopists’ relative unfamiliarity with the transnasal route. In conclusion, we have demonstrated that successful esophageal intubation is achieved more often GASTROINTESTINAL ENDOSCOPY
295
A Craig, J Hanlon, J Dent
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
by the transoral route for unsedated EGD with narrow-diameter endoscopes compared with the transnasal route. However, the fiberoptic endoscope can be used with equal efficacy by either route. The transoral route is preferred by endoscopists and avoids epistaxis. We believe that the transoral route should be the preferred route of insertion for unsedated EGD using the 5.9 mm videoendoscope and that either route can be used with the 5.3 mm fiberoptic endoscope. In addition, this study confirms previous data on patient preference and acceptable instrument performance for unsedated EGD with narrow-diameter endoscopes over conventional sedated EGD. As such, we recommend that unsedated EGD with narrow-diameter endoscopes be available to all patients undergoing diagnostic EGD. ACKNOWLEDGMENTS The following endoscopists contributed to the study: Dr. J. Argyrides, Dr. R. Fitch, Dr. R. Fraser, Dr. H. Harley, Dr. G. Hebbard, Dr. D. Hetzel, Dr. R.
296
GASTROINTESTINAL ENDOSCOPY
Holloway, and Dr. R. Johnson. We also thank L. Moorcraft, RN of our endoscopy unit nursing staff who assisted with the randomization of patients. REFERENCES 1. Shaker R. Unsedated trans-nasal pharyngoesophagogastroduodenoscopy (T-EGD): technique. Gastrointest Endosc 1994;40:346-8. 2. Nozaki R, Fujiyoshi T, Tamura M, Tsuchiya A, Takagi K, Takano M. Evaluation of small caliber transnasal panendoscopes for upper GI screening examination. Dig Endosc 1995;7:155-9. 3. Dean R, Dua K, Massey B, Berger W, Hogan W, Shaker R. A comparative study of unsedated transnasal esophagogastroduodenoscopy and conventional EGD. Gastrointest Endosc 1996;44:422-4. 4. Bampton P, Fitch R, Dent J. A comparison of transnasal and transoral oesophagogastroduodenoscopy. J Gastroenterol Hepatol 1998;13:579-84. 5. Zaman A, Hahn M, Hapke R, Knigge K, Fennerty M, Katon R. A randomized trial of peroral versus transnasal unsedated endoscopy using ultrathin video endoscope [abstract]. Gastrointest Endosc 1998;47:AB 61. 6. Lukomsky G, Ovchinnikov A, Bilal A. Complications of bronchoscopy. Chest 1981;79:316-21.
VOLUME 49, NO. 3, PART 1, 1999
Background: The aim of this study was to compare use and tolerance of transnasal and transoral diagnostic endoscopy with small-diameter endoscopes in unsedated patients. Methods: Patients being seen for diagnostic endoscopy were randomly assigned to have an unsedated transnasal or transoral procedure. Two prototype narrowdiameter endoscopes (a 5.3 mm fiberoptic endoscope and a 5.9 mm videoendoscope) were used. Results: Of 170 patients (64 women and 106 men) enrolled, 86 underwent transoral and 84 underwent transnasal endoscopy. The procedure was successfully completed using the transoral route in 85 of 86 patients and using the transnasal route in 74 of 84 patients (p = 0.004). Sixteen patients experienced mild epistaxis after transnasal endoscopy.The larger videoendoscope, when compared with the fiberoptic endoscope, accounted for a significantly higher proportion of failures (8 of 41 vs. 2 of 43, p = 0.046) and cases of epistaxis (12 of 33 vs. 4 of 41, p = 0.007). The time taken for transoral endoscopy was shorter than the transnasal route (oral mean 13.7 ± 0.5 minutes, nasal mean 15.2 ± 0.6 minutes, p = 0.054). There was no difference between the 2 groups with respect to tolerance of the procedure. From the endoscopist’s perspective, the only statistically significant difference between the 2 groups was that endoscope insertion was easier by the oral route (p = 0.007). Conclusions: Unsedated transnasal endoscopy with the videoendoscope was less successful compared with the transnasal fiberoptic instrument or when compared with either instrument passed transorally. Use of the larger diameter videoendoscope also resulted in significantly more epistaxis in the transnasal endoscopy group. Endoscopists find transoral introduction of the endoscope easier; this may reflect their relative unfamiliarity with the nasal route. Once intubation has been successfully achieved for either route, patient tolerance is the same. (Gastrointest Endosc 1999;49:292-6.) Esophagogastroduodenoscopy (EGD) is a commonly performed and well-established diagnostic procedure. Standard EGD involves oral passage of a flexible endoscope, 9 mm in diameter, after the oropharynx has been anesthetized with a local anesReceived July 22, 1998. For revision August 20, 1998. Accepted October 21, 1998. From the Department of Gastrointestinal Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia. The data were presented at the Poster Session at the American Society for Gastrointestinal Endoscopy Annual Meeting during DDW in New Orleans May 20, 1998. Published in abstract form in Gastrointestinal Endoscopy 1998;47:AB 28. Reprint requests: Mark Schoeman, PhD, Department of Gastrointestinal Medicine, Royal Adelaide Hospital. North Tce, Adelaide, South Australia, Australia 5000. Copyright © 1999 by the American Society for Gastrointestinal Endoscopy 0016-5107/99/$8.00 + 0 37/1/95331 292
GASTROINTESTINAL ENDOSCOPY
thetic spray or gargle. In Australia, intravenous sedation is given routinely to reduce patient discomfort and anxiety. However, sedation can lead to respiratory depression, and after the procedure patients require supervised recovery, which is sometimes prolonged. There are also significant indirect costs associated with the use of sedation because patients are unable to drive or to return to work for at least 12 hours after the procedure. Narrow-diameter endoscopes were developed in the early 1990s in the hope that the small caliber (< 6 mm) would reduce patient discomfort. The nasal route of insertion in particular was promoted because it was thought that this method would reduce pharyngeal discomfort and the risk of aspiration. Another potential advantage of nasal insertion is that patients are able to talk during the procedure, which can help reduce their anxiety. In addition, the VOLUME 49, NO. 3, PART 1, 1999
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
nasal route eliminates the risk that a patient will bite the endoscope. Initial studies in healthy volunteers 1 and in a cancer screening program in Japan 2 have shown that the transnasal unsedated procedure is safe and well tolerated, and that it allows adequate visualization of the upper gastrointestinal tract. Two studies directly compared unsedated transnasal EGD using a narrow diameter endoscope with standard sedated transoral EGD 3,4 and found acceptable patient tolerance and shortened recovery times with the first method. There is, however, some evidence from these earlier studies that the transnasal route is associated with an increase in adverse events such as epistaxis (13% in one study4), and it is our impression that some patients are unable to tolerate nasal intubation. In addition, endoscopists in our unit using the narrow-diameter endoscopes in unsedated patients have tended to choose the oral route of intubation because of its greater familiarity; patients are often unprepared to have a transnasal procedure. One study, published in abstract form only, directly compared the transnasal and the transoral routes of insertion for unsedated endoscopy with a narrowdiameter endoscope.5 This study, which included 50 patients, showed that the transoral route was better tolerated than the transnasal route. The aim of our study was to compare the transoral and the transnasal routes of insertion for unsedated EGD using narrow-diameter instruments in a large patient group, with respect to patient tolerance and endoscopists’ impressions. PATIENTS AND METHODS Entry into the trial was offered to all our patients aged 18 to 90 years referred for EGD to any of 8 endoscopists at our hospital from June 1996 to March 1998. After informed written consent was obtained, patients were randomly assigned to either transnasal or transoral endoscopy. We randomized the route of insertion using random numbers. The selected route was then sealed in an envelope, and the envelopes were sequentially numbered. After informed written consent was obtained, the next envelope was opened to determine the route of insertion. Patients were excluded from the study if there was a history of significant nasal trauma or surgery, if they were having a colonoscopy at the same appointment, if it was judged that a therapeutic procedure was likely to be required, or if there was a bleeding diathesis. Patients undergoing the transnasal procedure received a local anesthetic spray (lidocaine) to the nostril being intubated. Patients undergoing the transoral procedure received a local anesthetic gargle (amethocaine hydrochloride). Two prototype narrow-diameter endoscopes were used in the study (Olympus Australia, Oakleigh, Victoria): a 5.3 mm fiberoptic instrument (Fig. 1) and a 5.9 mm videoendoscope. Both endoscopes had a 2 mm accessory channel and VOLUME 49, NO. 3, PART 1, 1999
A Craig, J Hanlon, J Dent
Figure 1. Relative diameters of 3 endoscopes. Left, therapeutic endoscope, diameter 11 mm; middle, standard endoscope, diameter 9 mm; right, fiberoptic endoscope, 5.3 mm diameter. had similar performance characteristics. All endoscopists had prior experience with unsedated narrow-diameter endoscopy performed through either route. All procedures were performed with the patient in the left lateral position. Sedative medications were not routinely used, although atropine and prophylactic antibiotics could be used at the endoscopist’s discretion. An intravenous cannula was inserted only if atropine or antibiotics were given. All patients were monitored with pulse oximetry, and oxygen was administered if the oxygen saturation fell below 90%. The oxygen could be given via the nose or mouth depending on the route of endoscope insertion. After the procedure, the patient’s condition was monitored in a purpose-designed recovery room; fitness for discharge was assessed by using the modified recovery room American Pediatric Gross Assessment Record score. Data were collected from patients, the nursing staff and endoscopists. Specific data collected by the nursing staff included the time the patient entered and left the procedure room, the time spent in the recovery area before leaving, whether supplemental oxygen was required, whether any adverse event(s) occurred, and whether biopsies were performed. Patients were asked to score the procedure with visual analog scales. They scored how they tolerated the procedure, how well they generally felt, and how capable they felt about returning to work or normal activities. A visual analog scale was also used to assess the patient’s opinions on whether they would prefer an unsedated or sedated procedure if they required EGD in the future. Patients who had previously undergone a standard sedated procedure were asked whether they would want any future EGD performed with or without sedation. Endoscopist satisfaction was also assessed with visual analog scales for the level of difficulty encountered in intubation, for the adequacy of the view, for the performance of the endoscope with regard to air insufflation, the washing of the lens, and suction, for the degree of endoscope looping and for whether difficulty was encountered in entering the duodenum. If biopsy specimens were taken, the endoscopist was asked to rate the ease with which this task was performed. The study was reviewed and approved by the ethics committee of our hospital. Binary data were analyzed by means of chi-square tests of independence. Other data were analyzed by using GASTROINTESTINAL ENDOSCOPY
293
A Craig, J Hanlon, J Dent
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
Table 1. Procedure data Transoral
Transnasal
p Value
No. patients Age (years) Males Females Successful completion Procedure time (mean in minutes ± SE) Recovery time (mean in minutes ± SE) Adverse events
86 48.4 (19-84) 59 (69%) 27 (31%) 85 of 86 13.7 ± 0.5
84 51.1 (18-88) 47 (56%) 37 (44%) 74 of 84 15.2 ± 0.6
NS NS NS 0.004 0.054
10.0 ± 1.1
13.0 ± 1.4
NS
n=0
Supplemental O2 Biopsies
0 56
n=16 (epistaxis) 2 41
NS
Table 2. Mean patient response scores (out of 10 with 10 being the most positive response) ± standard error How did you tolerate the test? How do you feel? Could you return to work now? Would you be happy to have the test repeated in the form that you had it?
Transoral
Transnasal
p Value
6.4 ± 0.2
6.3 ± 0.3
NS
8.1 ± 0.3 8.6 ± 0.3
7.9 ± 0.3 8.4 ± 0.3
NS NS
9.0 ± 0.3
8.8 ± 0.3
NS
Patients who did not undergo a complete procedure were excluded. NS, Not significant
NS, Not significant
the Anova test. A p value less than 0.05 was taken as indicative of significant differences.
RESULTS The study included 170 patients (64 women, 106 men). The transoral route was used in 86 patients, and the transnasal route was used in 84 patients. The two groups were evenly matched for age (Table 1). The transoral group contained a higher proportion of men than did the transnasal group (Table 1), but this difference was not statistically significant. Similar numbers of patients underwent endoscopy with the fiberoptic endoscope and videoendoscope in each group (oral: video 42, fiberoptic 44; nasal: video 41, fiberoptic 43). The procedure was successfully completed in 85 of 86 patients by the transoral route and in 74 of 84 patients by the transnasal route (p = 0.004). Failed intubations in the transnasal group were caused by a combination of difficult insertion of the endoscope through the nasal passage and patient intolerance. Failed intubations occurred at a steady rate throughout the study period and were equally distributed among endoscopists. Patients who could not complete the procedure were excluded from any subsequent analysis. No complications were noted in the patients who had the transoral route, whereas 16 of 74 patients had epistaxis after the transnasal route. In 1 patient, the epistaxis continued for 8 hours but did not necessitate any specific intervention. In all other patients the epistaxis stopped within 5 minutes. No other adverse events were recorded. Two patients undergoing the transnasal route required supplemental oxygen for transient oxygen desaturation. The time taken for the endoscopy and the time spent in the recovery room are shown in Table 1. 294
GASTROINTESTINAL ENDOSCOPY
There was a trend toward a shorter procedure time for the transoral route. There was no difference between the two groups in the time spent in the recovery room. Patients in the transnasal group with epistaxis had a recovery time similar to that of the other patients in this group (mean 13.0 minutes). Patient assessments of the procedure are summarized in Table 2. There was no difference between the two groups with respect to any of the factors evaluated. Fifty-four patients had previously undergone a standard transoral sedated endoscopy, and 49 of these patients (oral: 25, nasal: 24) said that they would prefer to undergo an unsedated endoscopy with a narrow-diameter endoscope if required in the future. Endoscopist assessments are shown in Table 3. From the endoscopists’ perspective, the only statistically significant difference between the two groups was that endoscope insertion was judged to be easier by the oral route (p = 0.007). The difference between assessments of the adequacy of the endoscopic view almost reached statistical significance (p = 0.0501), with transoral being better than transnasal. With respect to ability to insufflate air, wash the lens, suction, instrument looping, and the ability to enter the duodenum there were no differences between the two groups. Biopsies were performed in 97 patients with no difference in the endoscopist score for performing biopsies (p = 0.7). All biopsies performed were adequate for histopathologic assessment. A subanalysis was performed on the video and fiberoptic instruments for the transnasal route and is summarized in Table 4. This disclosed a significantly lower rate of successful completion of the procedure with the videoendoscope (33 of 41 vs. 41 of VOLUME 49, NO. 3, PART 1, 1999
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
Table 3. Mean proceduralist scores (out of 10 with 10 being the most positive response) ± standard error Transoral Scope insertion View Insufflation Wash lens Suction Enter duodenum Looping
8.9 8.6 8.8 7.6 8.5 9.9 8.3
± ± ± ± ± ± ±
0.3 0.3 0.3 0.4 0.3 1.3 0.3
Transnasal 8.1 7.9 8.6 8.1 7.9 8.7 8.2
± ± ± ± ± ± ±
0.2 0.3 0.3 0.3 0.3 0.3 0.3
p value 0.007 0.0501 NS NS NS NS NS
43, p = 0.046) and a higher rate of epistaxis (12 of 33 vs. 4 of 41, p = 0.007). DISCUSSION This is the largest study that has specifically addressed the question of which route is best for insertion of a narrow-diameter endoscope in unsedated patients undergoing EGD. We found a higher rate of successful intubation with the transoral route. The transoral route was also the preferred route of insertion from the endoscopists’ perspective. One factor to account for these findings may be endoscopists’ relative unfamiliarity with the transnasal route, as the majority of routine endoscopic procedures are performed using the transoral route. Indeed, physicians who perform bronchoscopy would find a 10% failure rate for intubation with bronchoscopy unacceptable, although sedation is often used in this setting and bronchoscopes are still smaller than the narrow-diameter endoscopes. Once intubation had been successfully completed, patient tolerance was the same for both routes. The high rate of epistaxis (22.6%) may again be related to the endoscopists’ relative unfamiliarity with the transnasal route with resultant trauma to the nasal passage. Another explanation could be that the local anesthetic spray causes irritation. We did not use cocaine nasal spray, which is routinely used in otolaryngology and by chest physicians, as it has vasoconstrictive effects that probably reduce the incidence of bleeding. Epistaxis that is self-limited and does not necessitate intervention following transnasal bronchoscopy is reported as “common” but the incidence of significant epistaxis is low: 0.1% to 0.3%.6 All the episodes of epistaxis in this study resolved spontaneously. In the subanalysis of endoscope type, the failure of nasal intubation and the incidence of epistaxis was more common with the videoendoscope (5.9 mm) compared with the smaller fiberoptic (5.3 mm) VOLUME 49, NO. 3, PART 1, 1999
A Craig, J Hanlon, J Dent
Table 4. Sub-analysis of transnasal route between fiberoptic and video endoscopes Transnasal route Successful completion Epistaxis
Fiberoptic endoscope
Video endoscope
p Value
41 of 43 4 of 41
33 of 41 12 of 33
0.046 0.007
NS, Not significant.
endoscope. The fiberoptic instrument was passed with equal ease and with a comparable complication rate through either the mouth or the nose. This indicates that transnasal endoscopy should be performed only with endoscopes that are of a diameter of 5.3 mm or less. The larger videoendoscope, 5.9 mm in diameter, should be used only for unsedated transoral endoscopy. One advantage of unsedated transnasal endoscopy is a reduction in the risk of aspiration during the procedure compared with transoral endoscopy in sedated patients.1,3 During standard endoscopy, pharyngeal irritation may induce emesis which, when combined with oropharyngeal anesthesia, would predispose to aspiration. Transnasal endoscopy avoids these problems. In our study, there were no clinically evident aspiration events in either patient group. It is likely that a much larger study will be required to answer this question because the incidence of clinically evident and significant episodes of aspiration is low. It is our impression that patient sedation is the major contributor to occurrences of aspiration. Therefore, unsedated procedures are likely to be associated with a low risk of aspiration—patients are able to guard their airway more effectively. This study also allowed us to obtain further data that compare patient experiences of conventional sedated EGD with unsedated EGD performed with a narrow-diameter endoscope. We found that 90.7% of patients who had previously undergone sedated conventional EGD would prefer an unsedated EGD with a narrow-diameter endoscope in the future if required. Nozaki et al.2 reported a similar percentage of patients who found unsedated transnasal EGD with a small-diameter endoscope more comfortable than conventional sedated EGD. Our study also confirms that the performance characteristics of the narrow endoscopes are acceptable. The trend in this study toward better endoscopic views with the transoral route may be related to endoscopists’ relative unfamiliarity with the transnasal route. In conclusion, we have demonstrated that successful esophageal intubation is achieved more often GASTROINTESTINAL ENDOSCOPY
295
A Craig, J Hanlon, J Dent
Transnasal vs transoral unsedated endoscopy with small-diameter endoscopes
by the transoral route for unsedated EGD with narrow-diameter endoscopes compared with the transnasal route. However, the fiberoptic endoscope can be used with equal efficacy by either route. The transoral route is preferred by endoscopists and avoids epistaxis. We believe that the transoral route should be the preferred route of insertion for unsedated EGD using the 5.9 mm videoendoscope and that either route can be used with the 5.3 mm fiberoptic endoscope. In addition, this study confirms previous data on patient preference and acceptable instrument performance for unsedated EGD with narrow-diameter endoscopes over conventional sedated EGD. As such, we recommend that unsedated EGD with narrow-diameter endoscopes be available to all patients undergoing diagnostic EGD. ACKNOWLEDGMENTS The following endoscopists contributed to the study: Dr. J. Argyrides, Dr. R. Fitch, Dr. R. Fraser, Dr. H. Harley, Dr. G. Hebbard, Dr. D. Hetzel, Dr. R.
296
GASTROINTESTINAL ENDOSCOPY
Holloway, and Dr. R. Johnson. We also thank L. Moorcraft, RN of our endoscopy unit nursing staff who assisted with the randomization of patients. REFERENCES 1. Shaker R. Unsedated trans-nasal pharyngoesophagogastroduodenoscopy (T-EGD): technique. Gastrointest Endosc 1994;40:346-8. 2. Nozaki R, Fujiyoshi T, Tamura M, Tsuchiya A, Takagi K, Takano M. Evaluation of small caliber transnasal panendoscopes for upper GI screening examination. Dig Endosc 1995;7:155-9. 3. Dean R, Dua K, Massey B, Berger W, Hogan W, Shaker R. A comparative study of unsedated transnasal esophagogastroduodenoscopy and conventional EGD. Gastrointest Endosc 1996;44:422-4. 4. Bampton P, Fitch R, Dent J. A comparison of transnasal and transoral oesophagogastroduodenoscopy. J Gastroenterol Hepatol 1998;13:579-84. 5. Zaman A, Hahn M, Hapke R, Knigge K, Fennerty M, Katon R. A randomized trial of peroral versus transnasal unsedated endoscopy using ultrathin video endoscope [abstract]. Gastrointest Endosc 1998;47:AB 61. 6. Lukomsky G, Ovchinnikov A, Bilal A. Complications of bronchoscopy. Chest 1981;79:316-21.
VOLUME 49, NO. 3, PART 1, 1999