Colonic Polyps, Colorectal Neoplasm
Conditions
Keywords
Colonoscopy, Diagnostic Imaging, Colorectal Neoplasm, Primary Prevention, Cost and Cost Analysis, Endoscopy
Brief summary
The investigators hypothesize that high definition colonoscopy with close focus narrow band imaging features has a high diagnostic accuracy for colorectal polyp histology, and can replace formal pathologic assessment in cases of high diagnostic confidence.
Detailed description
A recent single arm prospective cohort study suggested that high definition colonoscopy with narrow band imaging is an acceptable strategy to diagnose polyp histology and determine future surveillance intervals. Endoscopic proficiency in macroscopic features to differentiate polyp histology can be attained in a relatively short time period. A new high definition colonoscope with close focus narrow band imaging features may further improve diagnostic accuracy for polyp differentiation, and can replace formal pathologic assessment in cases of high diagnostic confidence. Primary Aim: Compare the diagnostic accuracy of high definition narrow band imaging colonoscopy with and without close focus features for the macroscopic differentiation of neoplastic and non-neoplastic colorectal lesions using histopathologic diagnosis as the reference standard. Secondary Aims: Compare diagnostic characteristics of the colonoscopes, measure concordance of high confidence and accuracy, compare accuracy of predicted versus actual surveillance interval recommendations, perform a cost-effective analysis of endoscopic versus pathologic diagnoses, and measure complications.
Interventions
DEVICEClose focus HD NBI Colonoscopy System
Technically improved colonoscope with close focus high definition narrow band imaging. Optical specifications include a 2mm near field focal depth.
Sponsors
Kansas City Veteran Affairs Medical Center
VA Palo Alto Health Care System
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
DIAGNOSTIC
Masking
SINGLE (Subject)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Adult patients referred for routine colonoscopy
Exclusion criteria
* Known inflammatory bowel disease * Personal or family history of polyposis or non-polyposis syndrome * Presentation for emergency endoscopy * Inability to remove polyp due to coagulopathy or thrombocytopenia * Inability to provide informed consent
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Rate of Accurate High Confidence Polyp Histology Predictions by the Endoscopist in the Two Groups. | At the time of procedure | Measure of the percentage of accurate high confidence predictions by the endoscopist in the differentiation of neoplastic from non-neoplastic colorectal lesions, using the high definition NBI colonoscopy with and without close focus features. High confidence was assigned if the polyp had one or more features of Type 2 (neoplasia) or Type 1 (nonneoplasia) in the NICE classification and no features associated with the other histology Note: one patient may have multiple polyps. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cost | At the time of procedure | Measure the cost of colonoscopy with macroscopic histopathologic diagnosis of colorectal lesions compared to colonoscopy with conventional microscopic histopathologic diagnosis, on the lesions that were managed based on an accurate endoscopic diagnosis. A reduction in pathology specimens may improve the efficiency of the procedure and has direct pathology cost savings (as well as indirect savings, which were not measured). |
| Diagnostic Characteristics | At time of procedure | Compare the diagnostic characteristics (sensitivity, specificity, positive predictive value and negative predictive value) using the high definition narrow band imaging colonoscopy with and without close focus features. Accuracy: number of endoscopic predictions of adenomatous polyps histologically confirmed to be adenomatous/number of predicted hyperplastic polyps confirmed to be hyperplastic out of all polyps Sensitivity: number of endoscopic predictions (optical diagnosis) of adenomatous (neoplastic) polyps out of all histologically confirmed polyps Specificity: number of endoscopic predictions of hyperplastic (non-neoplastic) polyps out of all histologically confirmed polyps PPV: number of histologically confirmed adenomatous polyps out of all endoscopic predictions of adenomatous polyps NPV: number of histologically confirmed hyperplastic polyps out of all endoscopic predictions of hyperplastic (non-neoplastic) polyps Note: a patient may have multiple polyps |
| Accuracy of Predicted Versus Actual Surveillance Intervals | At the time of procedure | Compared the accuracy of predicted versus actual surveillance colonoscopy interval recommendations by determining number of patients with correct surveillance interval recommendation. |
| Learning Curve | At time of procedure. | Examine the impact of a learning curve (i.e. NPV of high confidence at each of endoscopist's first 50% of exams versus last 50% exams to endoscopically predict polyp histology). NPV is defined as number of histologically confirmed hyperplastic polyps out of all endoscopic predictions of hyperplastic (non-neoplastic) polyps. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Close Focus HD NBI Colonoscopy System Use of close focus to make optical diagnosis
Close focus HD NBI Colonoscopy System: Technically improved colonoscope with close focus high definition narrow band imaging. Optical specifications include a 2mm near field focal depth. | 277 |
| Current HD NBI Colonoscopy System Use of standard focus for optical diagnosis | 281 |
| Total | 558 |
Baseline characteristics
| Characteristic | Close Focus HD NBI Colonoscopy System | Current HD NBI Colonoscopy System | Total |
|---|---|---|---|
| Age, Continuous | 63 years STANDARD_DEVIATION 7.7 | 62.4 years STANDARD_DEVIATION 8.7 | 62.7 years STANDARD_DEVIATION 8.2 |
| Indication Screening | 104 Participants | 106 Participants | 210 Participants |
| Indication Surveillance | 121 Participants | 123 Participants | 244 Participants |
| Indication Symptoms | 52 Participants | 52 Participants | 104 Participants |
| Sex: Female, Male Female | 16 Participants | 12 Participants | 28 Participants |
| Sex: Female, Male Male | 261 Participants | 269 Participants | 530 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 277 | 0 / 281 |
| serious Total, serious adverse events | 0 / 277 | 0 / 281 |
Outcome results
Primary
Rate of Accurate High Confidence Polyp Histology Predictions by the Endoscopist in the Two Groups.
Measure of the percentage of accurate high confidence predictions by the endoscopist in the differentiation of neoplastic from non-neoplastic colorectal lesions, using the high definition NBI colonoscopy with and without close focus features. High confidence was assigned if the polyp had one or more features of Type 2 (neoplasia) or Type 1 (nonneoplasia) in the NICE classification and no features associated with the other histology Note: one patient may have multiple polyps.
Time frame: At the time of procedure
Population: We analyzed all diminutive sized polyps to assess the number of polyps that endoscopists were able to make an accurate high confidence optical diagnosis. High confidence was assigned if the polyp had one or more features of Type 2 (neoplasia) or Type 1 (non-neoplasia) in the NICE classification and no features associated with the other histology.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Close Focus HD NBI Colonoscopy System | Rate of Accurate High Confidence Polyp Histology Predictions by the Endoscopist in the Two Groups. | 75.3 Percent of accurate HC predictions |
| Current HD NBI Colonoscopy System | Rate of Accurate High Confidence Polyp Histology Predictions by the Endoscopist in the Two Groups. | 63.1 Percent of accurate HC predictions |
Secondary
Accuracy of Predicted Versus Actual Surveillance Intervals
Compared the accuracy of predicted versus actual surveillance colonoscopy interval recommendations by determining number of patients with correct surveillance interval recommendation.
Time frame: At the time of procedure
| Arm | Measure | Group | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Close Focus HD NBI Colonoscopy System | Accuracy of Predicted Versus Actual Surveillance Intervals | Actual (histopathological diagnosis) | 277 Participants |
| Close Focus HD NBI Colonoscopy System | Accuracy of Predicted Versus Actual Surveillance Intervals | Predicted (optical diagnosis) | 259 Participants |
| Current HD NBI Colonoscopy System | Accuracy of Predicted Versus Actual Surveillance Intervals | Actual (histopathological diagnosis) | 281 Participants |
| Current HD NBI Colonoscopy System | Accuracy of Predicted Versus Actual Surveillance Intervals | Predicted (optical diagnosis) | 259 Participants |
Secondary
Cost
Measure the cost of colonoscopy with macroscopic histopathologic diagnosis of colorectal lesions compared to colonoscopy with conventional microscopic histopathologic diagnosis, on the lesions that were managed based on an accurate endoscopic diagnosis. A reduction in pathology specimens may improve the efficiency of the procedure and has direct pathology cost savings (as well as indirect savings, which were not measured).
Time frame: At the time of procedure
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Close Focus HD NBI Colonoscopy System | Cost | 241 Specimen bottles that could be omitted |
| Current HD NBI Colonoscopy System | Cost | 0 Specimen bottles that could be omitted |
Secondary
Diagnostic Characteristics
Compare the diagnostic characteristics (sensitivity, specificity, positive predictive value and negative predictive value) using the high definition narrow band imaging colonoscopy with and without close focus features. Accuracy: number of endoscopic predictions of adenomatous polyps histologically confirmed to be adenomatous/number of predicted hyperplastic polyps confirmed to be hyperplastic out of all polyps Sensitivity: number of endoscopic predictions (optical diagnosis) of adenomatous (neoplastic) polyps out of all histologically confirmed polyps Specificity: number of endoscopic predictions of hyperplastic (non-neoplastic) polyps out of all histologically confirmed polyps PPV: number of histologically confirmed adenomatous polyps out of all endoscopic predictions of adenomatous polyps NPV: number of histologically confirmed hyperplastic polyps out of all endoscopic predictions of hyperplastic (non-neoplastic) polyps Note: a patient may have multiple polyps
Time frame: At time of procedure
Population: 975 total polyps were found with high confidence predictions made on 774 of them.We assessed diagnostic performance of the high confidence predictions. High confidence was assigned if the polyp had one or more features of Type 2 (neoplasia) or Type 1 (nonneoplasia) in the NICE classification and no features associated with the other histology.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Close Focus HD NBI Colonoscopy System | Diagnostic Characteristics | Accuracy | 88.5 percentage of high confidence prediction |
| Close Focus HD NBI Colonoscopy System | Diagnostic Characteristics | Sensitivity | 98.2 percentage of high confidence prediction |
| Close Focus HD NBI Colonoscopy System | Diagnostic Characteristics | Specificity | 73.9 percentage of high confidence prediction |
| Close Focus HD NBI Colonoscopy System | Diagnostic Characteristics | PPV | 85.0 percentage of high confidence prediction |
| Close Focus HD NBI Colonoscopy System | Diagnostic Characteristics | NPV | 96.4 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Diagnostic Characteristics | NPV | 92.0 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Diagnostic Characteristics | Accuracy | 87.0 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Diagnostic Characteristics | PPV | 84.4 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Diagnostic Characteristics | Sensitivity | 95.2 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Diagnostic Characteristics | Specificity | 75.7 percentage of high confidence prediction |
Secondary
Learning Curve
Examine the impact of a learning curve (i.e. NPV of high confidence at each of endoscopist's first 50% of exams versus last 50% exams to endoscopically predict polyp histology). NPV is defined as number of histologically confirmed hyperplastic polyps out of all endoscopic predictions of hyperplastic (non-neoplastic) polyps.
Time frame: At time of procedure.
Population: 975 total polyps were found and we were able to make high confidence predictions on 774 of them.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Close Focus HD NBI Colonoscopy System | Learning Curve | First half | 94.9 percentage of high confidence prediction |
| Close Focus HD NBI Colonoscopy System | Learning Curve | Second half | 96.7 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Learning Curve | First half | 88.0 percentage of high confidence prediction |
| Current HD NBI Colonoscopy System | Learning Curve | Second half | 95.8 percentage of high confidence prediction |
Post Hoc
Surveillance Colonoscopy Advanced Pathology Findings
Compare the advanced pathology and colon cancer incidence at surveillance colonoscopy of the groups who underwent high definition narrow band imaging colonoscopy with and without close focus features, and more specifically who underwent macroscopic versus microscopic polyp management. This applies if we bring the patient back after 5 years, we have not collected the data yet.
Time frame: 5 to 10 years after colonoscopy