Self-training in unsedated transnasal EGD by endoscopists competent in standard peroral EGD: prospective assessment of the learning curve

ORIGINAL ARTICLE: Clinical Endoscopy

Self-training in unsedated transnasal EGD by endoscopists competent in standard peroral EGD: prospective assessment of the learning curve ´ro ˆ me Dumortier, MD, PhD, Jean-Marc Dumonceau, MD, PhD Massimo Maffei, MD, Je Geneva, Switzerland

Background: Training programs in unsedated transnasal (UT) EGD are scarce. Objective: To prospectively assess the learning curve for unsupervised UT-EGD. Setting: Endoscopy service, without experience in UT-EGD. Subjects: Consecutive patients referred for diagnostic EGD. Intervention: UT-EGD was attempted in 140 study patients by 2 endoscopists who trained by themselves in UT-EGD (skilled endoscopist [n Z 70]; a trainee having recently achieved competency in conventional EGD [n Z 70]) and in 10 controls (endoscopist skilled in UT-EGD) by using a 4.9-mm-diameter videoendoscope. Main Outcome Measurements: Technical success, sedation administered, patient tolerance acceptance, procedure duration for each decade of 10 consecutive patients investigated by the same endoscopist; intention-to-treat analysis. Results: Both self-trained endoscopists fulfilled predefined criteria of competency in UT-EGD since the first attempts. They completed examinations of adequate quality with exclusive transnasal scope insertion (n Z 139 [99.3%]), no sedation (n Z 138 [98.6%]), and patient accepting repeat procedure (n Z 135 [96.4%]) in proportions not significantly different from controls for all decades. Compared with a median procedure duration of 5.5 minutes (interquartile range [IQR] 5.0-8.5 minutes) in controls, procedures were significantly longer for all trainee’s decades (eg, first decade 20.0 minutes [IQR 15.0-29.0 minutes], P!.001) but none for the skilled endoscopist. Overall discomfort, pain, gagging, and belching were not significantly different for study patients versus controls. Fifty-six of 69 study patients (81%) with a previous history of conventional EGD preferred UT-EGD. Limitations: Generalizability to other small-caliber endoscopes. Conclusions: Endoscopists competent in conventional EGD may obtain excellent results with UT-EGD (except for procedure duration) beginning with their first attempts, even without supervision or structured training. (Gastrointest Endosc 2008;67:410-8.)

Today’s standard for EGD in Western countries is peroral EGD with the patient under sedation.1-3 However, sedation presents several drawbacks, including additional costs (up to 40% of the total procedure costs) and risks (sedation has been responsible for a majority of endoscopy-related morbidity and mortality in several series).4-6 An alternative to sedated EGD, namely unsedated trans-

Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.07.024

nasal EGD (UT-EGD), has emerged with the development of small-caliber endoscopes. Compared with sedated peroral EGD, this technique offers similar or higher degrees of patient tolerance and acceptance with similar diagnostic yields.7-12 Small-caliber endoscopes may also be inserted perorally, but this approach does not seem to be associated with as good a patient tolerance as the transnasal approach.13-15 Despite specific advantages, UT-EGD has not been widely implemented.16 Endoscopists who have not adopted UT-EGD frequently cite a lack of training and doubts about the advantages of this technique as their major motivations. Specific training in UT-EGD may be required,

410 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 3 : 2008

www.giejournal.org

Abbreviations: IQR, interquartile range; UT, unsedated transnasal; VAS, visual analog scale.

Maffei et al

because most GI endoscopists are not familiar with many aspects of this technique, including the anatomy of nasal cavities and particularities of small-caliber endoscopes (eg, weak suctioning and air-inflation capabilities, presence of a single bending plane on the newest models). However, training programs in UT-EGD are scarce, and GI endoscopists may be reluctant to commit time and money in supervised training for an alternative to conventional peroral EGD.17 We prospectively assessed the learning curve of unsupervised training in UT-EGD by endoscopists who had no previous experience with this technique, including a trainee who recently achieved competency in conventional EGD and a skilled endoscopist. Results were compared with those obtained in identical conditions by an endoscopist with extensive experience in UT-EGD.

Unsupervised training in transnasal EGD

Capsule Summary What is already known on this topic d

Lack of training in unsedated transnasal EGD (UT-EGD) may impede the adoption of this technique, and few training programs are available.

What this study adds to our knowledge d

d

Endoscopists competent in conventional EGD who trained by themselves in UT-EGD fulfilled predefined criteria of competency in UT-EGD in their first attempts. Initial procedures were longer, but technical success, patient tolerance, and patient acceptance were as good as with an endoscopist skilled in UT-EGD.

Before study initiation, UT-EGD had never been performed at the University Hospital of Geneva. A 4.9mm-diameter videoendoscope specifically designed for UT-EGD and equipped with a single bending plane (GIF-N180; Olympus Optical Co [Europa], Hamburg, Germany) was obtained for the study purpose. The 2 ‘‘beginners’’ who participated in the study included (1) a trainee still involved in the learning process of standard GI endoscopy but who had recently achieved competency in upperGI endoscopy (number of upper-GI procedures performed before study initiation, 184),18 and (2) a skilled endoscopist with a 15-year experience in GI endoscopy. Beginners had no experience of any kind (eg, watching procedures) in UT-EGD or with endoscopes equipped with a single bending plane, but they had experience in unsedated standard peroral EGD (eg, among 100 consecutive diagnostic standard peroral GI endoscopy procedures performed immediately before study initiation, 50 and 41 procedures had been performed without sedation by the trainee and the skilled endoscopist, respectively). They educated themselves separately by (1) reading scientific material identified through PubMed searches, and (2) watching an educational movie that demonstrated nasal anesthesia and the performance of UT-EGD with the patient in the left decubitus position (Olympus). They performed UT-EGD without supervision, and procedure reports were closely monitored during the study by an endoscopist (J.D.) who had extensive experience in this technique (O1000 procedures). The 10 last patients included in the study had UT-EGD attempted by the endoscopist (J.D.) in conditions identical to the beginners’ conditions (endoscopy room, material, and nurses), and these served as controls.

2006 on the days when the beginners were assigned to cover routine EGD were considered eligible when the small-caliber endoscope was available. Exclusion criteria included a request for sedation by the patient before starting endoscopy, a previous history of upperdigestive surgery, severe bleeding diathesis (platelet count !50,000/mm3, prothrombin rate !50%), psychiatric diseases, age !18 years, allergy to lidocaine, previous inclusion in the study, nasal trauma, and an inability to give informed consent. The study was approved by the institutional ethics committee of the University Hospital of Geneva. Eligible patients were proposed to undergo UT-EGD after placement of an intravenous line (they were instructed that sedation could be given if they did not tolerate the examination and that if they chose not to enter the study, then the alternative was a standard peroral EGD under intravenous sedation with midazolam if they preferred); written informed consent for participation in the study was obtained in all cases. Details of the method used for UT-EGD in study patients varied over the study period. Most recently, patients were first asked to swallow 20 mL of diluted simethicone,19 and topical anesthesia was then performed by asking the patient to inhale 2% lidocaine (AstraZeneca, Stockholm, Sweden) and applying lidocaine by using a cotton-tipped swab, as previously described.7 The endoscope was then inserted under direct vision through the most patent nostril with the patient in the sitting position (if insertion failed through both nostrils, it was agreed to switch to the peroral route with the same instrument), and a standard EGD was performed, including retroflexion in the stomach (plus tissue sampling if deemed necessary). Patients could follow the procedure on an endoscopy screen and were encouraged to talk with the endoscopist during the examination.

Patients and intervention

Definitions

Consecutive patients referred to the University Hospital of Geneva for diagnostic EGD between June and November

Procedure durations were recorded from the beginning of topical anesthesia to complete endoscope withdrawal

www.giejournal.org

Volume 67, No. 3 : 2008 GASTROINTESTINAL ENDOSCOPY 411

PATIENTS AND METHODS

Unsupervised training in transnasal EGD

Maffei et al

(‘‘whole procedure duration’’) and from endoscope insertion to complete withdrawal (‘‘UT-EGD duration’’). Patients were discharged to the recovery room, and the diagnostic efficacy was rated by the examiner on a 10point visual analog scale (VAS) relative to conventional peroral endoscopy (0, terrible; 10, identical). The quality of examinations was rated by 1 endoscopist (J.D.) as adequate if satisfactory photographs were obtained from 4 locations (esophagus, fundus via retroflexion, antrum, and second portion of the duodenum). Upon discharge from the recovery room, patients were asked to self report on a questionnaire (1) their tolerance to the procedure (overall discomfort during the examination, pain, gagging, belching, and coughing) on the 10-point VAS (0, nonexistent; 10, unbearable),20 (2) if they would accept undergoing a repeat upper-GI endoscopy in identical conditions if medically indicated (yes or no), and (3) which modality (ie, ‘‘through the nose’’ or ‘‘through the mouth’’) they would prefer in case they had previously undergone conventional peroral GI endoscopy. If necessary, the patient filled the form with the help of an assistant who had not attended the procedure; patients were not informed that their answers would not be reviewed by their own endoscopist. Potential complications were systematically searched for and recorded at the time of patient discharge. Technical competency in UT-EGD was evaluated after each decade of 10 consecutive attempted examinations and was defined a priori (based on published data)11,12,19 as fulfillment of all of the 4 following conditions: (1) R90% successful transnasal scope insertion, (2) adequate examination quality in R90% of cases, (3) sedation required in %10% of patients, and (4) patient acceptance of undergoing the same procedure if medically indicated in R70% of cases. A particular procedure was defined as a successful UT-EGD if all of these conditions were fulfilled (ie, examination of adequate quality with the endoscope inserted transnasally, no sedation administered, and the patient accepting upper-GI endoscopy in identical conditions if medically indicated).

for the usual trainee without previous experience in upper-GI endoscopy to achieve competency in this technique (n Z 130),18 and the demonstration that additional endoscopy skills may be acquired with supervision limited to the first 10 procedures, with a plateau reached at 50 procedures in the case of EUS-guided FNA.17 All analyses were performed according to the intention-to-treat principle, including all attempted examinations. Results are expressed as the median value with the 25% to 75% interquartile range (IQR) or as a percentage. Comparisons were performed with the Pearson c2 or the Fisher exact test for categorical data and the Wilcoxon signed rank test for continuous variables. All tests were 2 sided, and P values less than .05 were considered statistically significant, except for comparisons with decades of patients where a P value less than .01 was chosen to adjust for multiple comparisons. Analyses were performed with JMP software (version 5.1.2, SAS Institute Inc, Cary, NC).

Study end points

Primary end point

The primary end point was to determine, for each beginner, the number of procedures required to achieve technical competency in UT-EGD. Predefined secondary end points included (1) the assessment of learning curves by comparing every decade of study patients versus controls for the following items: proportion of successful UT-EGDs, scores of patient tolerance, and procedure durations (UT-EGD and whole procedure), and (2) a comparison of these items for each series of 70 patients investigated by each beginner versus controls.

An arbitrary study sample of 70 patients per beginner was derived from the number of procedures required

Predefined criteria of technical competency in UTEGD were fulfilled by both beginners in all decades of patients, including the first one. Among the 20 patients included in the first decades, only one (5%) did not have a successful UT-EGD (defined as an examination of adequate quality with the endoscope inserted transnasally, no sedation administered, and the patient accepting a repeat procedure in identical conditions if medically indicated). In this patient (the second one investigated by the trainee, and the single study patient who did not have exclusive transnasal scope insertion), the pylorus could not be passed with the small-caliber endoscope, and a sedated conventional peroral EGD was performed.

412 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 3 : 2008

www.giejournal.org

Statistical analysis

RESULTS A total of 157 consecutive patients were assessed for eligibility, and UT-EGD was attempted by 2 beginners in 140 study patients followed by 10 control patients who were investigated by the endoscopist skilled in UT-EGD (Fig. 1). There were no significant differences between groups of patients with regard to demographics, indications, main findings at upper-GI endoscopy, as well as proportions of patients with biopsy sampling, but more specimens per patient were collected by the trainee compared with other endoscopists (P ! .0001) (Table 1). The most common endoscopic findings included gastroduodenal ulcers, esophagitis, and gastropathy of various types in 21 (14%), 16 (11%), and 14 patients (9%), respectively; endoscopy was reported as normal in 78 patients (52%).

Maffei et al

Unsupervised training in transnasal EGD

Figure 1. Flowchart showing patient inclusion in the study and groups of patients analyzed.

Secondary end points The assessment of learning curves mostly showed significantly longer procedure durations for the trainee compared with controls (Fig. 2). Durations of EGD and of the whole procedure (including topical anesthesia) remained significantly longer for all but 1 decade, and, during the trainee’s first decade, the mean EGD duration was more than 3 times that recorded for controls (20 minutes [IQR 15-29 minutes] vs 5.5 minutes [IQR 5-8.5 minutes]; P Z .0006). The proportion of successful UT-EGDs in study patients was at least equal to that observed in controls for all decades, and patient tolerance scores were not significantly different for any decade of study patients compared with controls (Fig. 3). In the comparison of each series of 70 study patients versus controls (Table 2), statistically significant differences were found for the trainee only: procedure durations were longer, more biopsy samples were collected, and less coughing was reported by patients investigated by the trainee versus controls. Seven of 140 procedures (5%) performed by beginners did not fulfill the criteria of a successful UT-EGD, mainly because 5 patients (4%) answered that they would not accept undergoing a repeat upper-GI endoscopy in identical conditions; unsuccessful UT-EGDs tended to be less frequent during the last 3 decades of study patients compared with previous ones (1/60 [2%] vs 6/80 [8%], respectively; www.giejournal.org

P Z .239). Fifty-six of the 69 study patients (81%) who had previously undergone a conventional peroral EGD indicated a preference for UT-EGD as performed by beginners.

Complications Complications were observed in 3 study patients (2%); they consisted of a self-limited epistaxis that required no particular intervention (n Z 2) and a vasovagal reflex with a transiently lowered blood pressure (n Z 1).

DISCUSSION Although a steep learning curve was anticipated based on reports of personal experience16,21 and, despite the lack of supervision, technical competency in UT-EGD was observed for both beginners since their first attempts. Our definition of technical competency reflected results similar to those reported in the literature in terms of completeness of endoscopic examination, patient tolerance to the procedure, patient acceptance of repeat examination, and requirement for sedation.10,11,19,20,22,23 Both beginners were competent in conventional peroral EGD before study initiation (very recently for the trainee), but none of them had practical knowledge of the anatomy of nasal cavities or had been exposed to UT-EGD other than through Volume 67, No. 3 : 2008 GASTROINTESTINAL ENDOSCOPY 413

Unsupervised training in transnasal EGD

Maffei et al

TABLE 1. Patients’ characteristics and endoscopic findings Beginners in UT-EGD

Median age (IQR) (y)

Trainee (N Z 70)

Skilled endoscopist (N Z 70)

Endoscopist skilled in UT-EGD (controls; N Z 10)

52.0 (37.8-73.3)

47.5 (36.0-61.3)

46.0 (27.8-73.8)

Male sex, no. (%)

35 (50)

34 (49)

5 (50)

Patients with previous EGD, no. (%)

34 (49)

35 (50)

4 (40)

Abdominal pain

11 (16)

25 (36)

0

Anemia

13 (19)

14 (20)

2 (20)

Workup before Roux-en-Y gastric by-pass

12 (17)

8 (11)

4 (40)

Dyspepsia

9 (13)

3 (4)

1 (10)

Portal hypertension

8 (11)

4 (6)

0

Exclusion of upper-GI bleeding

5 (7)

3 (4)

1 (10)

Heartburn

2 (3)

7 (10)

0

10 (14)

6 (9)

2 (20)

Patients with biopsy sampling, no. (%)

50 (71)

43 (61)

8 (80)

Median no. samples collected (IQR)

5.0 (4.0-6.0)y

3.0 (2.0-4.0)

Indication for endoscopy, no. (%)

Others* Biopsy sampling

2.0 (2.0-2.0)

Main endoscopic finding in each patient, no. (%) Gastroduodenal ulcer

8 (11)

10 (14)

3 (30)

Esophagitis (including Barrett’s esophagus)

8 (11)

7 (10)

1 (10)

Gastropathy

5 (7)

8 (11)

1 (10)

Gastric polyps

2 (3)

3 (4)

0

Othersz

5 (7)

11 (16)

0

Normal

42 (60)

31 (44)

5 (50)

*Including chronic diarrhea, screening for esophageal cancer, and follow-up of peptic ulcer disease. yP ! .0001. zIncluding hiatal hernia, esophageal candidosis, esophageal varices, esophageal and gastric cancers, and celiac disease.

personal research of published material and educational movies. Procedure duration was the single factor that was affected by the learning process: procedure lengthening (vs controls) was not significant for the skilled endoscopist, but it was major (up to 3 and 5 times that of controls and of conventional peroral EGD, respectively) and sustained up to study termination for the trainee (although it shortened during the first 4 decades).20 We can only speculate about the reasons why such long durations did not translate into reduced patient tolerance: scope insertion, the most uncomfortable part of the procedure (as with nasogastric

suction tubes), was probably not longer than for controls (it was not measured), and patients were encouraged by the trainee to speak during UT-EGD. Procedure duration was not included in our a priori definition of technical competency in UT-EGD (otherwise the trainee would not have achieved competency before study termination), because this criterion has not been proposed by the American Society for Gastrointestinal Endoscopy to assess competency, and slow procedure performance is common during learning processes.19,24-26 However, this may have an impact on endoscopy schedules and costs (together with the higher number of biopsy samples collected by the trainee), which

414 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 3 : 2008

www.giejournal.org

Maffei et al

Unsupervised training in transnasal EGD

should not be overlooked. Other parameters recorded with beginners compared favorably with those observed in controls or reported in the literature, even when the first examinations were included in the analysis: (1) transnasal endoscope insertion was successful in all patients (previous reports 79%-100%),20,23 (2) 96% of patients answered that they would accept a repeat examination (previous reports 69%-95%),10,19 (3) sedation was administered in 2% of patients (previous reports 0%-6%),11,27 (4) 81% of patients preferred the transnasal over the conventional peroral approach (previous reports 71%-91%),10,27 and (5) complications were detected in 2% of patients (previous reports 0%-37%).22,28

Factors that may have contributed to these good results include the familiarity of beginners with endoscopy procedures in unsedated patients (and hence with special attention required to technical details, such as topical anesthesia, patient reassurance during the procedure, and avoidance of air overinsufflation),16,21 and the use of a new, very slim, videoendoscope. This videoendoscope (GIF-N180; Olympus) was specifically designed for UTEGD (the ‘‘N’’ stands for nasal) and presents a 4.9-mmdiameter, 110-cm-long insertion tube with a tip flexing up/down (in a single plane) at 210 /120 . We anticipated that the availability of a single bending plane would be challenging, but we think that the reduction in diameter allowed by this modification compared with other models of GI endoscopes, indeed facilitated the procedures (the absence of lateral bending plane was compensated by torque maneuvers without particular difficulty). Randomized controlled trials recently showed that a diameter of the insertion tube %5.3 mm is associated with higher success of transnasal intubation, patient tolerance and acceptance of the procedure, and lower morbidity.22,29,30 The longer insertion tube and higher degree of tip deflexion of this endoscope compared with previous models from the same manufacturer might also have facilitated duodenal intubation and inspection of the gastroesophageal junction on retroflexion.21 Therefore, it is uncertain whether our results may be generalized to other, less slim models of small-caliber endoscopes. Other limitations of our study include a potential bias for reporting (patients were not informed that their own endoscopist would not be reviewing their answers to the questionnaire) and for recalling (item ‘‘preference for transnasal or conventional peroral approach’’), but similar results regarding patient preference for the transnasal approach were obtained by using an identical methodology.10,15 We elected to perform even the first procedures without supervision, because, in our opinion, this reflects clinical practice, in particular, the scarce availability of education programs in UT-EGD and time/money constraints of GI endoscopists.17,18 The absence of supervision seemed acceptable because the need for training was not certain, sedation could be administered if patients did not tolerate the procedure (all of them had an intravenous line), and results were continuously monitored by the endoscopist experienced in UT-EGD. Given the results, it is doubtful whether supervision would have improved technical success or patient tolerance. However, it cannot be excluded that lesions were missed and expert observation of skill might, therefore, be useful, particularly if this procedure is being taught to trainees who have not yet reached competency in standard peroral GI endoscopy. Nevertheless, our results do suggest that one of the major reasons cited by GI endoscopists for not adopting UT-EGD, ie, the lack of training, is unfounded. Technical changes were made by beginners after inclusion of about 50 patients; these included (1) replacement

www.giejournal.org

Volume 67, No. 3 : 2008 GASTROINTESTINAL ENDOSCOPY 415

Figure 2. Median durations from endoscope insertion to complete endoscope withdrawal (UT-EGD duration) and from beginning of topical anesthesia to complete endoscope withdrawal (whole procedure duration) for each decade of 10 consecutive study patients investigated by each beginner (dotted lines) and for 10 controls (horizontal line). Vertical lines indicate 25% to 75% IQR, and signs indicate decades, with a median duration significantly different from controls ()P !.01; ))P !.005; )))P ! .001).

Unsupervised training in transnasal EGD

Maffei et al

Figure 3. For each decade of 10 consecutive study patients investigated by each beginner and for 10 controls, bars in the left upper graph indicate the percentage of successful UT-EGDs (ie, examinations of adequate quality with the endoscope inserted transnasally, no sedation administered, and the patient accepting to undergo a repeat procedure), and boxes in other graphs represent the 25% to 75% IQR of patients’ symptoms on 10-point VAS (0, nonexistent; 10, unbearable), with horizontal lines representing median values.

of cocaine spray for nasal anesthesia by 2% lidocaine (administered via aspiration through the nostril and application with a cotton-tipped swab), which seemed easier and more reliable; (2) having the patient swallow, 5 minutes before the endoscopy, diluted simethicone (20 mL) to decrease the prevalence of foam; (3) performance of EGD with the patient in a seated position (rather than the left lateral position, which was initially used), in front of an endoscopy screen if the patient desired (to favor communication about endoscopic findings); (4) manual air inflation with a syringe or a hand bulb pump connected to the biopsy channel of the endoscope, which was quicker than the air pump of the endoscopy tower; and (5) use of a 5F biopsy forceps with elongated cuffs (FB-231K; Olympus) to take larger biopsy samples than

those provided with the model that was at our disposal for radial echoendoscopes (FB-19K-1; Olympus). Coughing was less frequently reported by study patients than by controls. This finding might be related to the use of lidocaine in pump spray (Xylocaine; AstraZeneca), which is known to be irritating,31 to perform topical anesthesia in controls. Unfamiliarity of the endoscopist skilled at UT-EGD with the setting of our endoscopy suite may also account for some differences observed between study patients and controls (patients investigated by the skilled endoscopist in his service reported coughing similar to study patients investigated by beginners).29 Some questions remain unanswered by this study and should be addressed in future prospective trials, including (1) the assessment of the ability to sample an intended

416 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 3 : 2008

www.giejournal.org

Maffei et al

Unsupervised training in transnasal EGD

TABLE 2. Endoscopy characteristics, patient tolerance and complications Beginners in UT-EGD Trainee (N Z 70) Successful UT-EGD, no. (%)

Skilled endoscopist (N Z 70)

Endoscopist skilled in UT-EGD (controls; N Z 10)

66 (94)

67 (96)

8 (80)

Exclusive transnasal scope insertion

69 (99)

70 (100)

9 (90)

No sedation administered

68 (97)

70 (100)

9 (90)

Adequate examination quality*

69 (99)

70 (100)

10 (100)

Patient accepting repeat procedure

68 (97)

67 (96)

8 (80)

Criteria used to define successful UT-EGD, no. (%)

Median duration (IQR) (min) Endoscopy

10.0 (10.0-13.3)y

7.0 (6.0-8.0)

5.5 (5.0-8.5)

Whole procedure

17.0 (15.0-21.0)y

11.0 (9.0-13.0)

12.0 (9.8-13.5)

10.0 (9.0-10.0)

10.0 (9.0-10.0)

9.0 (9.0-10.0)

Overall discomfort

2.0 (1.0-3.0)

2.0 (0.8-4.1)

1.0 (0.8-6.5)

Pain

1.0 (0-2.3)

1.2 (0-4.0)

1.0 (0-2.5)

Gagging

0 (0-2.0)

1.0 (0-2.5)

1.5 (0-6.8)

Belching

0 (0-2.0)

1.0 (0-2.0)

1.0 (0-5.0)

Coughing

0 (0-1.0)z

0 (0-1.0)

1.0 (0-2.5)

Median diagnostic efficacy (IQR) (endoscopist assessment on VAS; 0, terrible; 10, identical to conventional EGD) Patient tolerance (VAS; 0, nonexistent; 10, unbearable)

Patients preferring UT- EGD vs conventional EGD, no. (%) Complications, no. (%)

27/34 (79)

29/35 (83)

2/4 (50)

2 (3)

1 (2)

1 (10)

Epistaxis

1 (1)

1 (1)

1 (10)

Vasovagal reaction

1 (1)

0

0

*Based on photographs from the esophagus, fundus via retroflexion, antrum, and second portion of the duodenum. yP value in comparison with controls: P ! .0001. zP value in comparison with controls: P ! .01.

area during the first cases, particularly in the esophagus and the upper part of the greater curvature, with a single bending plane endoscope; (2) the possibility that learning curves may be relatively different for various individuals, because only 1 trainee was included in this study;32 and (3) the acquisition of competency by trainees who have not yet reached competency in standard upper-GI endoscopy (this would require assessment of other aspects of competency in UT-EGD, which are not specific to the transnasal approach, including indications, informed consent, and alternatives).18,33 In summary, we found that endoscopists who had no experience in UT-EGD and who were not supervised were technically competent in this technique from their first attempts. Procedure duration was the only parameter affected by the learning process, and its lengthening

(compared with procedures performed by an endoscopist skilled in UT-EGD) was marginal for a skilled endoscopist and more significant for a trainee who had recently achieved competency in conventional EGD. This had no impact on examination quality, patient tolerance, and acceptance of repeat procedures, which were in the upper range of previous reports.

www.giejournal.org

Volume 67, No. 3 : 2008 GASTROINTESTINAL ENDOSCOPY 417

ACKNOWLEDGMENTS We thank Olympus Europe (Hamburg, Germany) for the loan, at no cost, of an ultrathin GIF-N180 videoendoscope for the study duration (Olympus Europe had no role in the collection, analysis, and interpretation of data, or in the decision to seek publication); we are

Unsupervised training in transnasal EGD

indebted to the nurses from the University Hospital in Geneva, Switzerland, who made this study possible. DISCLOSURE The authors report that there are no disclosures relevant to this publication. Olympus provided a small-caliber endoscope (GIF-N180) at no cost for this research. Olympus had no other role, in particular, in the collection, analysis, and interpretation of data, or in the decision to seek publication.

REFERENCES 1. Heuss LT, Froehlich F, Beglinger C. Changing patterns of sedation and monitoring practice during endoscopy: results of a nationwide survey in Switzerland. Endoscopy 2005;37:161-6. 2. Cohen LB, Wecsler JS, Gaetano JN, et al. Endoscopic sedation in the United States: results from a nationwide survey. Am J Gastroenterol 2006;101:967-74. 3. Faulx AL, Vela S, Das A, et al. The changing landscape of practice patterns regarding unsedated endoscopy and propofol use: a national Web survey. Gastrointest Endosc 2005;62:9-15. 4. Abraham NS, Fallone CA, Mayrand S, et al. Sedation versus no sedation in the performance of diagnostic upper gastrointestinal endoscopy: a Canadian randomized controlled cost-outcome study. Am J Gastroenterol 2004;99:1692-9. 5. Mokhashi MS, Hawes RH. Struggling toward easier endoscopy. Gastrointest Endosc 1998;48:432-40. 6. Quine MA, Bell GD, McCloy RF, et al. Prospective audit of upper gastrointestinal endoscopy in two regions of England: safety, staffing, and sedation methods. Gut 1995;36:462-7. 7. Saeian K, Staff DM, Vasilopoulos S, et al. Unsedated transnasal endoscopy accurately detects Barrett’s metaplasia and dysplasia. Gastrointest Endosc 2002;56:472-8. 8. Saeian K, Staff D, Knox J, et al. Unsedated transnasal endoscopy: a new technique for accurately detecting and grading esophageal varices in cirrhotic patients. Am J Gastroenterol 2002;97:2246-9. 9. Saeian K, Townsend WF, Rochling FA, et al. Unsedated transnasal EGD: an alternative approach to conventional esophagogastroduodenoscopy for documenting Helicobacter pylori eradication. Gastrointest Endosc 1999;49:297-301. 10. Dumortier J, Napoleon B, Hedelius F, et al. Unsedated transnasal EGD in daily practice: results with 1100 consecutive patients. Gastrointest Endosc 2003;57:198-204. 11. Preiss C, Charton JP, Schumacher B, et al. A randomized trial of unsedated transnasal small-caliber esophagogastroduodenoscopy (EGD) versus peroral small-caliber EGD versus conventional EGD. Endoscopy 2003;35:641-6. 12. Thota PN, Zuccaro G Jr, Vargo JJ 2nd, et al. A randomized prospective trial comparing unsedated esophagoscopy via transnasal and transoral routes using a 4-mm video endoscope with conventional endoscopy with sedation. Endoscopy 2005;37:559-65. 13. Dumortier J, Ponchon T, Scoazec JY, et al. Prospective evaluation of transnasal esophagogastroduodenoscopy: feasibility and study on performance and tolerance. Gastrointest Endosc 1999;49:285-91. 14. Murata A, Akahoshi K, Sumida Y, et al. Prospective randomized trial of transnasal versus peroral endoscopy using an ultrathin videoendoscope in unsedated patients. J Gastroenterol Hepatol 2007;22:482-5. 15. Yagi J, Adachi K, Arima N, et al. A prospective randomized comparative study on the safety and tolerability of transnasal esophagogastroduodenoscopy. Endoscopy 2005;37:1226-31.

418 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 3 : 2008

Maffei et al 16. Saeian K, Shaker R. Unsedated ultrathin EGD. Gastrointest Endosc 2006;64:874-6. 17. Mertz H, Gautam S. The learning curve for EUS-guided FNA of pancreatic cancer. Gastrointest Endosc 2004;59:33-7. 18. American Society for Gastrointestinal Endoscopy. Principles of training in gastrointestinal endoscopy. From the ASGE. Gastrointest Endosc 1999;49:845-53. 19. Zaman A, Hapke R, Sahagun G, et al. Unsedated peroral endoscopy with a video ultrathin endoscope: patient acceptance, tolerance, and diagnostic accuracy. Am J Gastroenterol 1998;93:1260-3. 20. Birkner B, Fritz N, Schatke W, et al. A prospective randomized comparison of unsedated ultrathin versus standard esophagogastroduodenoscopy in routine outpatient gastroenterology practice: does it work better through the nose? Endoscopy 2003;35:647-51. 21. Shaker R. A wake-up call? Unsedated versus conventional esophagogastroduodenoscopy. Gastroenterology 1999;117:1492-5. 22. Craig A, Hanlon J, Dent J, et al. A comparison of transnasal and transoral endoscopy with small-diameter endoscopes in unsedated patients. Gastrointest Endosc 1999;49:292-6. 23. Dean R, Dua K, Massey B, et al. A comparative study of unsedated transnasal esophagogastroduodenoscopy and conventional EGD. Gastrointest Endosc 1996;44:422-4. 24. Choi IJ, Kim CG, Chang HJ, et al. The learning curve for EMR with circumferential mucosal incision in treating intramucosal gastric neoplasm. Gastrointest Endosc 2005;62:860-5. 25. Iacopini G, Frontespezi S, Vitale MA, et al. Routine ileoscopy at colonoscopy: a prospective evaluation of learning curve and skill-keeping line. Gastrointest Endosc 2006;63:250-6. 26. Mehdizadeh S, Ross A, Gerson L, et al. What is the learning curve associated with double-balloon enteroscopy? Technical details and early experience in 6 U.S. tertiary care centers. Gastrointest Endosc 2006;64: 740-50. 27. Jobe BA, Hunter JG, Chang EY, et al. Office-based unsedated smallcaliber endoscopy is equivalent to conventional sedated endoscopy in screening and surveillance for Barrett’s esophagus: a randomized and blinded comparison. Am J Gastroenterol 2006;101: 2693-703. 28. Alami RS, Schuster R, Friedland S, et al. Transnasal small-caliber esophagogastroduodenoscopy for preoperative evaluation of the high-risk morbidly obese patient. Surg Endosc 2007;21:758-60. 29. Dumortier J, Josso C, Roman S, et al. Prospective evaluation of a new ultrathin one-plane bending videoendoscope for transnasal EGD: a comparative study on performance and tolerance. Gastrointest Endosc 2007;66:13-9. 30. Horiuchi A, Nakayama Y. Unsedated ultrathin EGD by using a 5.2-mmdiameter videoscope: evaluation of acceptability and diagnostic accuracy. Gastrointest Endosc 2006;64:868-73. 31. Abe K, Miyaoka M. Trial of transnasal esophagogastroduodenoscopy. Dig Endosc 2006;18:212-7. 32. Marshall JB. Technical proficiency of trainees performing colonoscopy: a learning curve. Gastrointest Endosc 1995;42:287-91. 33. Cass OW, Freeman ML, Peine CJ, et al. Objective evaluation of endoscopy skills during training. Ann Intern Med 1993;118:40-4.

Received May 7, 2007. Accepted July 9, 2007. Current affiliations: Service of Gastroenterology and Hepatology (M.M., J.-M.D.), Geneva University Hospitals, Geneva, Switzerland, Department of Digestive Diseases (J.D.), E. Herriot Hospital, Lyon, France. Reprint requests: Jean-Marc Dumonceau, MD, Division of Gastroenterology and Hepatology, Geneva University Hospitals, Micheli-du-Crest St, 24, 1205 Geneva, Switzerland.

www.giejournal.org