| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Mayo Clinic, 200 First St. SW, Charlton 2, Rochester,
MN 55905.
2 Present address: Department of Radiology, Carle Clinic, Urbana, IL.
3 Department of Radiology, Mayo Clinic, Scottsdale, AZ.
4 Department of Internal Medicine, Mayo Clinic, Rochester, MN.
Received June 18, 2007;
accepted after revision April 23, 2008.
Address correspondence to D. M. Hough
(Hough.David{at}mayo.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Two hundred sixty-two patients (74 men, 188 women; mean age, 64 years; range, 21–92 years) consecutively referred for same-day or next-day CT colonography after incomplete colonoscopy underwent low-dose diagnostic CT before rectal tube insertion and CO2 insufflation.
RESULTS. Perforation was found on the low-dose CT scans of two of the 262 patients (0.8%; 95% CI, 0.1–2.7%). One of these patients had no symptoms; the other had mild abdominal discomfort at the time of CT.
CONCLUSION. The rate of occult colonic perforation after incomplete colonoscopy may be significant. For patients referred for CT colonography after incomplete endoscopy, use of low-dose diagnostic CT before rectal tube insertion and insufflation is indicated.
Keywords: colon • colonography • colonoscopy • CT colonography • perforation
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Perforation is a recognized complication of colonoscopy; reported perforation rates range from one case in 3,115 procedures (0.032%) to one case in 510 procedures (0.196%) [12–15]. Very rare reports of colonic perforation at CTC have emphasized that patients with obstructive lesions are at increased risk of perforation [16, 17] and that approximately one half of patients with colonic perforation as a result of CTC do not have symptoms [16, 18, 19]. The short time between incomplete colonoscopy and same-day or next-day CTC may not be adequate to allow some perforations to become clinically apparent. Because of the risk of exacerbating a clinically unsuspected perforation during insufflation at CTC, which can increase the risk of sepsis, screening for the presence of extraluminal gas before insufflation for CTC may be of benefit. The aims of this study were to establish that low-dose CT of the abdomen and pelvis can be used to detect colonic perforation in patients referred for CTC after incomplete colonoscopy and to estimate the prevalence of clinically occult perforation among these patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Institutional review board approval was obtained for the study, and only patients who had consented to the use of their medical records for investigational purposes were included. Informed consent was not needed because the low-dose scans were obtained for the patients' clinical benefit; that is, the low-dose scans could be used to avoid rectal tube insertion and colonic inflation that might have exacerbated a preexisting perforation.
The total effective radiation dose for low-dose CT colonography was calculated with organ-dose weighting factors established by the International Commission on Radiation Protection [20], scanner-specific Monte Carlo organ dose coefficients [21], and dose-calculation software [22]. This method entails energy spectrum–specific computer simulated x-ray interactions in a mathematic human model to determine normalized radiation doses to various organs. The organ doses are scaled according to the prescribed CT parameters, consideration being given to measured x-ray tube output values. The organ doses are then weighted and summed to yield the total effective dose.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of the patients with perforation was a 61-year-old woman undergoing screening colonoscopy. The colonoscope was introduced to the proximal sigmoid colon but because of tortuosity of the colon was not passed farther. The patient was referred for same-day CTC. She did not report symptoms to the endoscopy or radiology nurses before CT. The low-dose scans revealed unanticipated pneumatosis and a small amount of mesenteric air in the pelvis (Fig. 1A, 1B). On further questioning, the patient reported that she had experienced nausea before and after the colonoscopy but otherwise maintained that she had no symptoms. Physical examination revealed no abdominal tenderness or peritonism. The patient was admitted overnight and treated with IV fluids and antibiotics. She was discharged the next day in good condition and was given a course of oral antibiotics. The subsequent course was uneventful.
|
|
The second patient with perforation was a 70-year-old woman who had undergone screening colonoscopy. The colonoscope was not advanced more than 25 cm because of acute angulation of the sigmoid colon. After approximately 5 minutes of maneuvering, the colonoscope was removed, and a smaller-caliber endoscope was introduced. A possible mucosal tear was identified, but it was unclear to the endoscopist whether perforation had occurred. The patient was referred for same-day CTC. The low-dose CT scan obtained before insufflation showed extensive retroperitoneal gas (Fig. 2A, 2B, 2C). On questioning after the CT examination, the patient said she had mild lower abdominal discomfort. A physical examination revealed mild tenderness in the lower abdomen but no peritonism. The patient was admitted to the hospital and treated with IV fluids and antibiotics. During the night, worsening abdominal pain and a 38°C fever developed. A diatrizoate meglumine and diatrizoate sodium solution enema examination the following morning showed no evidence of leak. The symptoms resolved over the next 48 hours, and the patient was discharged taking oral antibiotics after 3 days in the hospital.
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that low-dose CT can depict clinically occult perforation before the patient undergoes CTC for incomplete colonoscopy. Two perforations were found in 262 patients (0.8%; 95% CI, 0.1–2.7%). One of these perforations was clinically unsuspected, and the other was suspected because of suspicion of a mucosal tear at endoscopy, even though the patient was generally free of symptoms. This second patient may have presented with delayed perforation at a later time if same-day CTC had not been performed. These data suggest that clinically occult perforation after colonoscopy may be more common than anticipated, especially after failed or incomplete colonoscopy. Our study numbers were small, however, and the CI for the true prevalence of clinically occult perforation after incomplete colonoscopy is wide.
Studies of overall perforation rates after all colonoscopic procedures indicate much lower rates than reported in this study, in the range of one in 3,115 procedures (0.032%) to one in 510 procedures (0.196%) [12–15]. Most of these perforations are clinically apparent after both successful and incomplete procedures. Risk factors for colonic perforation at colonoscopy are similar to the risk factors for incomplete colonoscopy, including colonic stricture and obstruction, advanced diverticulosis, tortuosity of the colon, previous abdominal or pelvic surgery, female sex, and advanced patient age [12, 23–26]. Thus it can be anticipated that the perforation rates observed in our study would be higher than for routine complete colonoscopy. In addition, endoscopists do not routinely order imaging for diagnosis of occult perforation after colonoscopy, so it is to be expected that asymptomatic perforations are underreported in the literature. To our knowledge, there have been no reports of studies addressing the perforation rate at incomplete colonoscopy or the rates of clinically unsuspected perforation.
Perforation at colonoscopy is most often caused by mechanical trauma by the colonoscope but also can be caused by polypectomy, electrocautery, and, in rare instances, barotrauma from overinflation [26–28]. Mechanical injury is the most common cause of perforation, and the sigmoid colon is the most frequent site [27, 28]. Both of the perforations in our study were caused by mechanical trauma in the sigmoid colon. Dissection of gas into the bowel wall, mesentery, and extraperitoneal soft tissues accounted for the appearance of extraluminal gas on the CT scans we obtained. Pneumoperitoneum is commonly found after symptomatic colonoscopic perforation [27], but we did not encounter any cases of pneumoperitoneum.
We began performing low-dose CT on all patients referred for CTC after incomplete colonoscopy in June 2004. Low-dose CT was preferred over upright abdominal radiography and digital CT scout imaging because it has greater sensitivity for detection of small volumes of retroperitoneal and intramural air and because of greater convenience to the patient. Upright radiography of the abdomen would be sensitive for pneumoperitoneum but would probably not depict small amounts of extraperitoneal air. Extraperitoneal air from occult perforation is likely to be common and underreported, and the risks of exacerbating a perforation during gas insufflation at CTC are high. The total effective dose for the low-dose CT scan is approximately one fifth the dose of single-pass standard CTC, or the equivalent of approximately 10 chest radiographic examinations. (The dose reduction of the low-dose technique relative to that of the standard technique is not exactly equal to the ratio of their tube current–exposure time products.) We therefore believe that the benefits of low-dose CT to diagnose occult perforation after incomplete colonoscopy outweigh the risk of the small radiation dose to the patient.
It is advantageous to the patient that CTC be performed the same day as or the day after incomplete colonoscopy because the patient does not need to repeat the bowel preparation. This protocol also avoids unnecessary delay in patient care and may decrease patient anxiety caused by waiting days or weeks for CTC. It also ensures that the structural evaluation of the colorectum is complete. If a clinically unsuspected perforation occurs at colonoscopy, it is likely to be small and may not be clinically relevant in most cases. Burling et al. [16] reported that all four patients with asymptomatic colonic perforation in their CTC series recovered without hospitalization or medical therapy. However, in patients with clinically unsuspected perforation at colonoscopy followed by same-day or next-day CTC, a small perforation can be exacerbated and result in clinically significant bowel perforation, increasing the risk of peritonitis or retroperitoneal sepsis. If CTC were to be performed at least a few days after colonoscopy, a diagnostic scan before insufflation might not be necessary because a small subclinical perforation would be expected to heal. However, the time needed for this healing is unknown. In addition, any extraluminal air would be expected to resorb within a few days, so a scan to search for free air may not be of value.
Extra caution may be needed in examinations of patients who have undergone biopsy at colonoscopy, especially if an obstructing lesion is the reason for incomplete colonoscopy. It has been found in a pig model [29] that the intraluminal pressure needed to perforate a normal colon does not differ from that needed to perforate a colon that has been subjected to superficial biopsy (excluding the muscularis propria). The colon, however, can perforate at lower pressures after deep biopsy that includes the muscularis propria. If gas is insufflated at CTC after such a biopsy, the increased volume during insufflation can be accommodated by the rest of the colon, but if an obstructing lesion is present, only a short segment of distal colon may be present to accommodate the increased intraluminal volume leading to increased pressure and perforation. There is no good way for a radiologist to determine whether a biopsy has been deep or superficial. Therefore, if biopsy or polypectomy has been performed and the patient has a stricture proximal to the biopsy site, it may be prudent to wait 1–2 weeks or more before performing CTC.
Incomplete colonoscopy reached the level of the rectosigmoid in 80 of our study patients, and both perforations were in patients from this group, giving a perforation rate of 2.5% in this subset of patients. Colonic intubation proximal to the sigmoid was achieved in 70% of the patients in our study, and no perforations were found in these patients. Although the numbers are small, these findings suggest that patients in whom the colonoscope cannot be passed beyond the sigmoid colon may be at greater risk of perforation than are other patients.
A limitation of our study was that we did not perform additional follow-up of the patients to determine whether they had delayed perforation not apparent at low-dose CT. However, we would expect that any tiny perforation not resulting in findings of extraluminal gas would be unlikely to be exacerbated by the subsequent CO2 insufflation for CTC. This assumption is somewhat validated by the fact that in none of the 260 patients with normal findings at low-dose CT was perforation found during CTC. Moreover, follow-up of these 260 patients would not have told us whether delayed perforation resulted from the initial colonoscopy or the subsequent CTC.
Although the goal is to avoid exacerbation of preexisting perforation in patients referred for incomplete endoscopy, most patients undergoing CTC for incomplete endoscopy do so safely, even in the presence of an obstructing lesion. In a study [10] (confirmed by personal communication, January 2008) with 546 patients who underwent CTC after incomplete endoscopy, there were no reported perforations despite a high incidence of obstructing tumors (7.5%), diverticulitis (14%), excessive colonic spasm (26%), and redundant or tortuous loops (39.9%). The reason for the discrepancy in our study is unknown, but we believe we found sufficient reason to perform low-dose CT as a precautionary measure before completing the full structural examination with CTC.
Colonic perforation after colonoscopy can be clinically occult. We believe that our findings justify performing low-dose diagnostic CT before rectal tube insertion and gas insufflation in all patients referred for same-day or next-day CTC after incomplete colonscopy in order to minimize the risks associated with exacerbating perforation.
Acknowledgments
We thank Kay L. Egner for assistance with the data collection.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
