INTRODUCTION — Variations in anatomy can make colonoscopy and small bowel enteroscopy technically difficult to perform and to teach. In addition, it is not always possible to precisely localize lesions within the bowel, and lesion localization is important if patients require subsequent endoscopic or surgical therapy.
One method for overcoming these limitations is to provide a view of the endoscope during the procedure. This was initially done with fluoroscopic guidance, but this approach was never routinely used in clinical practice because of its inconvenience, added expense, and necessity for radiation exposure. An alternative to fluoroscopy is magnetic endoscopic imaging (MEI), which provides real-time three-dimensional views of the endoscope shaft configuration and its location within the abdomen. This may aid trainees and less experienced endoscopists as they perform routine procedures and may aid experienced endoscopists during difficult cases.
Issues related to MEI will be reviewed here. Other options for localizing lesions within the colon are discussed separately. (See "Tattooing and other methods for localizing gastrointestinal lesions".)
CLINICAL APPLICATIONS — The ability to visualize the shaft of the scope during colonoscopy and enteroscopy may be helpful in multiple settings. It may aid the endoscopist during a difficult procedure, it may help with lesion localization, and it may facilitate teaching endoscopic technique.
Technically challenging colonoscopy — Factors that can increase the difficulty of colonoscopy include variations in colonic anatomy, loop formation during instrument insertion, and the presence of abdominal or pelvic adhesions. Increased colonic mobility can predispose to loop formation, whereas adhesions can interfere with normal scope insertion and loop reduction.
Influence of anatomy — The sigmoid and transverse colon are both mobile on a mesentery or "mesocolon," which can be affected by manipulations of the colonoscope. Most colonoscope looping occurs within these segments. Another factor that predisposes to loop formation is that the transverse colon of female patients (and, less commonly, male patients) may extend down into the pelvis [1].
Although the descending colon and ascending colon are usually relatively fixed in the left and right paravertebral gutters, respectively, these portions of the colon are mobile in a significant minority of patients. As an example, there is no fixation at the splenic flexure in approximately 20 percent of patients, which allows atypical colonoscope looping in the left colon [2].
On the other hand, the normally mobile sigmoid colon may have decreased mobility due to sigmoid adhesions that can result from diverticulitis, endometriosis, or pelvic surgery.
Looping and adhesions — Several studies have looked at the causes of difficult colonoscopy. In a series of 500 patients in whom fluoroscopy was used, 16 percent of cases performed by experienced colonoscopists were categorized as difficult examinations (defined as no advancement of the colonoscope tip for at least five minutes) [3]. This was due to recurrent looping in 80 percent and to sigmoid adhesions in the remainder. The colonoscopists were frequently incorrect in identifying the site of looping, and in 30 percent of examinations, they were mistaken as to the location of the tip of the colonoscope.
Looping also makes the examination more difficult for patients as pain episodes may correspond to looping seen on MEI [4]. The pain is greatest with looping in the sigmoid colon, particularly in women and in those with irritable bowel syndrome.
Other risk factors — A number of other risk factors have been described for difficult colonoscopy:
●On barium enema, there is a correlation between technically difficult colonoscopy and a long transverse colon or sigmoid colon adhesions [1,5]. These factors in part explain the higher rate of difficult colonoscopy in women compared with men [1,5-7]. In a review of 2194 colonoscopies performed by a single experienced colonoscopist, the respective rates of difficult colonoscopy in women and men were 31 and 16 percent [7].
●Colonoscopy is more likely to be difficult in older women with diverticular disease and adhesions and in men with constipation and a long colon. Computed tomography (CT) colonography data confirm that "difficult to colonoscope" patients tend to have longer colons (particularly in the sigmoid and transverse colon), have more acute bends, and are more likely to be female [6].
Enteroscopy — Negotiating the small bowel with an endoscope is challenging due to the mobile nature and length of the small bowel. Various technologies facilitate deep access into the small bowel, such as balloon-assisted enteroscopy and spiral endoscopy, but these techniques are technically challenging. In a small study of 10 patients who underwent single balloon enteroscopy, use of an MEI probe was shown to be safe and was associated with accurate detection and control of loops with good concordance with fluoroscopy [8]. (See "Overview of deep small bowel enteroscopy".)
Lesion localization — Precise localization of lesions within the colon is particularly important when surgical resection is required for a malignancy or a lesion needs to be reinspected at a later date. However, even skilled colonoscopists can misjudge the position of the colonoscope tip. It is easy to become confused about whether the tip is in the proximal sigmoid colon or at the splenic flexure, and the midtransverse colon or hepatic flexure may be misidentified as being the cecum.
While a number of visual clues are commonly used to determine the location of the tip of the colonoscope, they are often misleading. As examples, the triangular appearance of the folds that are considered to be typical of the transverse colon may also be seen in the descending colon, while the bluish hue associated with the hepatic flexure may also be seen at the splenic flexure (especially within a long sigmoid loop).
Other clues such as the length of instrument inserted, transillumination through the anterior abdominal wall, abdominal palpation, and the responsiveness and feel of the instrument can also be misleading.
Imprecise localization of a lesion can result in incorrect preoperative information being given to the surgeon [9]. One role for MEI may be in localizing cancers seen at colonoscopy. A study found that MEI achieved 94 percent accuracy in determining the location of 82 cancers that could be reached endoscopically [10]. This compared favorably with abdominopelvic computed tomography (CT), which localized 83 percent of cancers. Current guidelines recommend tattooing lesions for easy identification during follow-up examinations or surgery, though this may not occur if cancer was not suspected at the time of the colonoscopy and polypectomy. If, during the initial colonoscopy, MEI was used to document the location of the lesion, the confidence regarding the location may be greater than if it was only localized based on endoscopic findings. (See "Tattooing and other methods for localizing gastrointestinal lesions".)
COLONOSCOPY TRAINING — Colonoscopy training relies heavily upon apprenticeship. However, such training can be frustrating for both the trainee and the instructor. The trainee can often find it difficult to appreciate why certain maneuvers are indicated or effective; the instructor may be unable to assess why the trainee is "stuck" without taking over the procedure, leaving the trainee to act as an observer.
Fluoroscopy and magnetic endoscopic imaging help to address these frustrations because they allow the instructor to guide the trainee without having to take over the procedure.
ROLE FOR FLUOROSCOPY — The problems associated with performing colonoscopy provide a rationale for the use of some imaging modality during colonoscopy. Fluoroscopy increases confidence in lesion localization, aids with loop reduction during difficult examinations, and can be a useful tool when teaching [6,11,12].
However, fluoroscopy is not routinely used for multiple reasons [13]:
●Fluoroscopy can add time and expense to colonoscopy, exposes patients and staff to radiation, and is not always available.
●The area visible on-screen covers only a small part of the abdomen and may therefore give a misleading assessment of complex loops, especially when the patient lies in the conventional left lateral position.
●The two-dimensional radiological image does not provide an impression of the three-dimensional configuration of the colonoscope shaft loops. As a result, it is limited in its ability to guide rotational maneuvers used for straightening loops.
●Because of radiation hazards, fluoroscopy cannot be used to confirm that the assistant's hand is appropriately placed when applying pressure to the patient's abdomen to control loops.
Magnetic endoscopic imaging is an alternative to fluoroscopy that addresses these limitations.
MAGNETIC ENDOSCOPIC IMAGING
General principles — With MEI, electromagnetic fields are sensed by small coils within the colonoscope and larger ones alongside the patient, producing a moving image of the colonoscope configuration on a display monitor (movie 1 and picture 1 and picture 2) [14].
There are two commercially available MEI systems for colonoscopy; ScopeGuide (Olympus Corporation, Tokyo), and ScopePilot (Pentax Medical, Hamburg).
The first-generation ScopeGuide consisted of a large standalone unit with a receiver dish, processing unit, and display screen that is positioned alongside the patient. This was connected to a dedicated colonoscope containing built-in transmitter coils. A coil catheter can be inserted through the instrument/suction channel of a standard colonoscope, although this is less ideal because the catheter prevents suctioning and is too fragile for routine use. The transmitter coils were incorporated into a dedicated MEI colonoscope, resulting in a bulkier-than-standard colonoscope. The second-generation ScopeGuide (UPD-3) has a much smaller receiver dish (picture 2) with the image displayed as a picture-in-picture (PIP) on the main screen that is displaying the endoscope view (picture 1). The processing unit sits with the main endoscope image processing unit. The most recent commercial endoscope (Olympus 290) series has incorporated the transmitter coils in both standard and slim insertion tube colonoscopes.
ScopePilot provides similar ability to image the shape and position of the colonoscope. It also utilizes a picture-in-picture view for displaying the image of the scope configuration.
Image display — Software calculates the three-dimensional (3D) position and orientation of each receiver coil, and the data are displayed in real-time as a computer-rendered 3D image of the colonoscope shaft configuration (movie 1).
With ScopeGuide, the shaft configuration can be displayed either in anteroposterior (AP) view alone or in split-screen view, combining AP and lateral views side by side. The split-screen view helps to clarify the colonoscope loop configuration in 3D (figure 1). The monitor can also be configured to maintain the AP view despite changes in patient positioning. With ScopePilot, the image can be rotated in both the horizontal and vertical axis.
Hand marker — Targeted abdominal pressure applied by an assistant can be useful in managing colonoscope loops. Using an external coil held in the assistant's hand, it is possible to determine the position of the hand relative to the colonoscope, as well as the effect of the applied pressure on looping (figure 1). This feature is available for both ScopeGuide and ScopePilot.
IMPACT OF MEI — MEI helps with loop reduction, application of abdominal pressure, localization of the tip of the colonoscope, and providing instruction in colonoscopy. It may also have a place in competency assessment during the credentialing process for colonoscopy.
Loop identification and reduction — Correct loop identification is required for a successful reduction of looping during colonoscopy. However, endoscopists frequently miscategorize loops, especially if they are complex or unusual.
In a study of 100 consecutive colonoscopies done with MEI by experienced colonoscopists who were blinded to the imager view, various looping configurations were seen [15]. Despite skilled and careful insertion technique, loops occurred in 91 percent of patients (table 1).
The endoscopists incorrectly identified the loops in 69 percent of cases, with less common looping configurations, such as counterclockwise spiral loops, being incorrectly assessed in 100 percent. In addition, when abdominal pressure was applied, it was generally inaccurate, due either to hand misplacement away from the apex of the loop or to inaccessibility of the loop deep within the abdomen.
There are numerous loop configurations that may occur during colonoscopy (figure 2 and figure 3) [16]. The following examples illustrate how MEI may aid with loop reduction:
●An "N" loop as seen from the front or AP view has been said to be the most common sigmoid looping configuration (figure 2). However, many supposed N loops have a spiral component. MEI can help identify this configuration, which requires the endoscopist to "push through" the descending colon to the splenic flexure before straightening the colonoscope with pull-back and clockwise torque.
●A "reversed alpha" loop consists of a counterclockwise spiral (figure 2). An endoscopist without MEI often does not realize it is counterclockwise and is therefore unable to remove the loop with conventional clockwise torque. Once recognized, pulling back with strong counterclockwise torque straightens the loop. Counterclockwise torque is then applied during further insertion to prevent the sigmoid loop from reforming.
Tip localization — MEI aids with localization of the colonoscope tip, which may be important for confirming cecal intubation and lesion localization. Contrast studies following MEI-guided application of endoclips in predefined anatomical locations showed good correlation between MEI-defined and actual anatomical clip locations [17-20]. However, it is possible to be misled by the shape of the instrument alone. As a result, the typical features of the cecum and ileocecal valve should always be confirmed before concluding that a colonoscopy is complete.
Colonoscopy performance and training — A number of randomized trials have assessed the impact of MEI on colonoscopy performance with mixed results [18,21-28]. Some studies demonstrated improvements in parameters such as cecal intubation rates [21,22], loop reduction [21,23,29], external pressure application [21,27], pain scores [22], and insertion times [21,24]. However, improvements in loop reduction and/or cecal intubation rates did not always result in decreased procedure times or sedation requirements [21,23,25,29,30]. One direct comparison with double balloon endoscopy found the use of MEI with a transparent cap was less effective for achieving caecal intubation within 30 minutes in patients with previously incomplete or difficult colonoscopy [26].
In a meta-analysis of 13 studies, the use of MEI resulted in a lower risk of failed cecal intubation compared with conventional colonoscopy (6.8 versus 10.1 percent; risk difference 4 percent, 95% CI 0 to 7 percent) and slightly shorter time to cecal intubation (mean difference 0.6 minutes, 95% CI 0.3-0.9 minutes) [31]. However, when the analysis was restricted to high-quality studies, the differences were no longer seen.
The impact of MEI on colonoscopy performance appears to be greatest for trainees and inexperienced colonoscopists [21,24,32,33]. However, there is also evidence of some benefit from MEI for experienced colonoscopists [18,21,30,31,34].
●In a pilot trial, a single beginner colonoscopist (15 previous colonoscopies) performed procedures under supervision [32]. Patients were randomly assigned to have their procedures with or without MEI. Improved loop management was noted when MEI was employed. A plateau was reached after 50 cases, when 90 percent completion rates were achieved. Once that plateau was reached, no difference was seen comparing cases performed with and without MEI.
●A trial of 296 colonoscopies looked at the effect of MEI on colonoscopy performance by a trainee (200 previous cases) and by two experienced endoscopists (>5000 previous cases) [21]. The trainee, when using MEI, had significant improvements in cecal intubation rates, insertion times, duration of colonoscope looping, number of straightening attempts, and accuracy of hand pressure. The experienced endoscopists had similar but less marked benefits. MEI did aid with cecal intubation in eight difficult examinations, although there was no effect on patient pain scores or sedation requirements.
●A trial of 152 consecutive colonoscopies performed by 10 staff surgeons compared the utility of MEI with fluoroscopy. Examination time was significantly shorter in the MEI group, while avoiding radiation exposure, and completion rates were similar, but low, in both groups [35].
●There is also evidence that using simulated MEI during training on a virtual reality simulator can produce training outcomes for novice colonoscopists, equivalent to those of one-to-one expert bedside teaching.
●A multicenter randomized controlled trial demonstrated that the performance of novices who had 16 hours of training on the colonoscopy simulator matched the performance of those with equivalent time of standard patient-based colonoscopy training [36].
●A study of three inexperienced trainees found that a half-day training session on a model using MEI improved performance scores, increased cecal intubation rates, and reduced intubation times and analgesia requirements over 86 colonoscopies [24].
Credentialing for colonoscopy — More robust and accurate methods are now being developed to assess competency at colonoscopy. Studies have demonstrated that MEI can be useful for rating competence at colonoscopy [37] and that the use of simulated MEI may have a place in the credentialing process [38]. MEI is now used consistently as part of the competency assessment during accreditation to perform bowel cancer screening colonoscopy in the United Kingdom.
SAFETY — The risks of colonoscopy with MEI are the same as those associated with any colonoscopy and include adverse reactions to sedation, perforation, and bleeding. There is minimal radiation exposure related to MEI due to the field generation method that is used. The strength of the fields produced by the Olympus imaging system is significantly lower than the fields considered acceptable for continuous public exposure. The total radiation exposure is less than that generated by a television monitor.
There was a theoretical safety concern that exposure to magnetic fields could be harmful. This concern was based upon limited data suggesting that exposure to magnetic fields arising from overhead power lines or cellular phones could have adverse effects. However, it is the electric fields, and not the magnetic fields, around power lines or phones that are of actual concern.
Studies of the use of MEI in patients with implantable cardiac devices have not shown any interference or change in settings, and are considered safe [39].
SUMMARY AND RECOMMENDATIONS
●Imaging during colonoscopy or small bowel enteroscopy may be helpful in multiple settings. It may aid the endoscopist during a difficult procedure, it may help with lesion localization, and it may facilitate teaching endoscopic techniques. (See 'Clinical applications' above and 'Impact of MEI' above.)
●Magnetic endoscopic imaging (MEI) operates by sensing electromagnetic fields using small coils within the colonoscope and larger ones alongside the patient, producing a moving image of the colonoscope configuration on a display monitor. (See 'Magnetic endoscopic imaging' above.)
●The use of MEI helps with loop reduction, application of abdominal pressure, and localization of the tip of the colonoscope. It has been shown to be useful for trainees learning colonoscopy and may have a place in the credentialing process. (See 'Impact of MEI' above.)
ACKNOWLEDGMENT — The author and UpToDate thank Brian P Saunders, MD, MRCP, and Christopher B Williams, BM, FRCP, FRCS, who contributed to earlier versions of this topic review
