multidetector ct of the post- operative colon: review of normal

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

515GASTROINTESTINAL IMAGING

Stefanie Weinstein, MD • Samuel Osei-Bonsu, MD • Rizwan Aslam, MBChB • Judy Yee, MD

If not properly recognized, the normal postoperative appearance of the pelvis following colorectal surgery can be misinterpreted as disease, in-cluding infection or recurrent tumor. However, multidetector computed tomography (CT) with the supplemental use of multiplanar reformation clearly demonstrates the expected postoperative anatomic changes in this setting. The high-resolution images achievable with multidetector CT enable the radiologist to play an important role in the postoperative as-sessment of patients following colon surgery. Whenever possible, the ra-diologist should be aware of the specific indication for the study, the type of surgery that was performed (ranging from segmental bowel excision to more extensive radical resection), and what anastomoses were created. This knowledge, as well as familiarity with the normal multidetector CT appearances of various postoperative complications, is critical for prompt diagnosis and appropriate management of these complications and for better differentiation of complications from normal findings. ©RSNA, 2013 • radiographics.rsna.org

Multidetector CT of the Post-operative Colon: Review of Normal Appearances and Common Complications1

Abbreviations: APR = abdominoperineal resection, IBD = inflammatory bowel disease, IPAA = ileal pouch–anal anastomosis

RadioGraphics 2013; 33:515–532 • Published online 10.1148/rg.332125723 • Content Codes: 1From the Department of Radiology, San Francisco VA Medical Center, UCSF Medical Center, 4150 Clement St (114), San Francisco, CA 94121. Received April 5, 2012; revision requested May 29 and received July 6; accepted September 12. For this journal-based SA-CME activity, the authors, editor, and reviewers have no relevant relationships to disclose. Address correspondence to S.W. (e-mail: [email protected]).

©RSNA, 2013

SA-CME

See www.rsna .org/education

/search/RG

LEARNING OBJECTIVES FOR TEST 4

After completing this journal-based SA-

CME activity, partic-ipants will be able to:

■ Discuss the indica-tions for the most common colorectal surgical procedures and the differences between these pro-cedures.

■ Describe normal findings in the post-operative colon.

■ Differentiate ex-pected postoperative findings from post-surgical complica-tions.

516 March-April 2013 radiographics.rsna.org

Figure 1. CT image obtained in a 68-year-old man who had undergone APR for rectal cancer shows a mottled presacral collection (arrow) representing air and packing material.

IntroductionColorectal surgery is a common procedure that is performed for a variety of benign and malig-nant colonic diseases, including inflammatory bowel disease (IBD), diverticulitis, and colorectal cancer. The choice of surgical technique is influ-enced by the severity of disease and the extent of disease spread in the bowel and, to some extent, by the preference and experience of the surgeon (1,2). Recent improved understanding of the cause and progression of various colorectal dis-eases has led to substantial advances in surgical management (3). Current surgical procedures include segmental bowel excision, sphincter-spar-ing techniques, and extensive radical resection. Similar to other surgical procedures, colorectal surgery is frequently associated with postopera-tive complications (1,3,4). Technologic advance-ments in multidetector computed tomography (CT) have allowed accurate assessment of the expected anatomic changes in the postoperative colon, as well as evaluation of postoperative com-plications (3–10).

In this article, we review the most common colorectal surgical procedures, including abdomi-noperineal resection (APR), anterior resection, the Hartmann procedure, restorative proctocolec-tomy with ileal pouch–anal anastomosis (IPAA), and segmental resection, and describe their ex-pected postoperative appearances at multidetec-tor CT. In addition, we describe the imaging fea-tures of the early postoperative complications that are most commonly seen after colorectal surgery, including wound complications, anastomotic leak, fistula, intraperitoneal abscess, pouchitis, small bowel obstruction and pseudo-obstruction, and diversion colitis.

Common Procedures and Expected Findings

Abdominoperineal ResectionAPR, also referred to as the Miles procedure, in-volves excision of a portion of the sigmoid colon and of the entire rectum and anus with use of an abdominal-perineal approach. The abdominal component consists of excision of the distal colon and mobilization of the rectum, with creation of a permanent end colostomy, usually in the left lower abdomen through the rectus abdominis muscles.

The perineal component consists of excision of the anus and distal rectum with removal of the mobilized abdominal bowel (11). APR is com-monly indicated for (a) low-lying rectal cancers (either primary or recurrent) less than 8 cm above the pectinate line (1), (b) anal malignancies, and (c) anorectal complications of IBD (12). A post-operative pelvic cavity is created at proctectomy. This open defect is obliterated with either (a) use of drains to eliminate accumulated fluid or blood, or (b) packing using the mobilized pedicle of the greater omentum. Alternatively, the open defect can be closed with a reapproximation of the pelvic peritoneum by means of primary closure of the perineal incision, or reconstructed with a myocu-taneous flap (rectus abdominis, gracilis, or gluteal muscle) or a biologic mesh (Fig 1) (13). With development of both sphincter-sparing low ante-rior resection and the circular stapling technique, along with improvements in adjuvant cancer therapy such as radiation therapy, APRs are being performed less frequently. However, despite the higher incidence of perineal complications associ-ated with APR (eg, wound infections [13]) and the lack of sphincteric preservation (14), APR is often still the ideal procedure for anorectal carcinoma with extensive sphincter involvement and end-stage IBD (12).

The extensive excision performed in APR causes significant alteration in the abdomino-

RG • Volume 33 Number 2 Weinstein et al 517

Figure 4. (a) CT image obtained in a 65-year-old man who had undergone APR shows asymmetric nodular soft tissue (arrows) anterior to the sacrum, a finding that represents normal posteriorly displaced seminal vesicles but was mistaken for lymphadenopathy. (b) CT image obtained in a different patient who had undergone APR shows presacral soft tissue (arrow), a finding that represents normal posteriorly displaced seminal vesicles.

Figure 3. CT image obtained in a 50-year-old man who had undergone APR for rectal cancer shows an elongated bladder (white arrow), a posterior location of the prostate gland (black arrow), and presacral soft tissue (arrowhead), all of which are expected early postoperative findings.

Figure 2. CT image obtained in a 77-year-old man who had undergone APR for complications of Crohn disease demonstrates residual presacral soft-tissue fi-brosis (white arrow) posterior to the seminal vesicles (black arrow). Follow-up CT performed 2 years later showed no change.

pelvic anatomy. Multidetector CT demonstrates the repositioning of this anatomy in the space previously occupied by the excised viscera (Fig 2). The bladder, seminal vesicles, and small bowel typically move posteriorly into a precoc-cygeal location (Fig 3) (15). In addition, the uterus often drops into the presacral space, particularly if the pelvic peritoneal lining is not reconstructed (10). A presacral soft-tissue mass

often represents the newly displaced pelvic structures, postsurgical fibrosis, or postsurgi-cal granulation tissue (Fig 4). The mass can be clearly identified on axial images, but its relationship to other pelvic structures and bony anatomy is better delineated with multidetector CT supplemented with multiplanar reformation.


Figure 5. (a, b) Axial (a) and sagittal (b) CT images obtained in a patient who had undergone APR 5 days earlier demonstrate the expected changes, with a myocutaneous flap extending to the perineum from the rectus abdominis muscle (white arrows) and a fluid- and air-containing collection in the pre-sacral space (black arrows). (c, d) Follow-up axial CT images obtained 2 years later show a marked decrease in the amount of presacral fluid and resolution of the air but persistent presacral soft tissue (black arrow in c). The perineal flap is again noted (white arrow in c). There is an end colostomy (ar-rowhead in d) and an atrophic right rectus abdominis muscle from the prior flap harvest (arrow in d).

Difficulty has been reported in distinguishing the characteristic finding of postoperative presa-cral soft tissue from tumor recurrence (15–17). However, evolution of these changes on serial CT images helps in differentiation (Fig 5). The

majority of patients who undergo APR have pre-sacral midline masses with a maximum diameter of 3–5 cm at initial evaluation. The postopera-tive soft-tissue mass typically demonstrates some decrease in size and change in configuration in


Figure 6. CT (top) and fused PET/CT (bottom) images obtained in a patient who had undergone APR show a heterogeneous presacral mass (arrow on CT image) without radiotracer activity, a finding that is consistent with known fibroid uterus.

the months following surgery. Masses typically have ill-defined margins at initial studies but more distinct borders at follow-up. The change in size at follow-up can vary from minimal to marked, with only a thin sliver of residual tis-sue persisting. At other times, the presacral soft tissue remains stable and can persist for up to 2 years after surgery, or even indefinitely. In contrast, a soft-tissue mass due to recurrent dis-ease almost always grows and typically becomes more infiltrative and ill defined over time. Lack of growth over 1–2 years, clinical stability, and normal carcinoembryonic antigen levels are additional evidence that the mass represents normal postoperative changes (10). Causes of false-positive findings of tumor recurrence in-clude uterine fibroids, fibrosis, and postoperative abscess; multiplanar reformatted images can be particularly useful in distinguishing among these entities. Sagittal reformatted images are particu-

larly helpful in ascertaining the position of the uterus and the relationship of the postoperative soft tissue to other pelvic structures. Because of the occasional difficulty in distinguishing this soft tissue from tumor recurrence, a baseline CT evaluation is performed within 2–4 months after surgery, and it is recommended that follow-up CT be performed every 6 months thereafter to exclude recurrent tumor in oncology patients (1). In a study by Even-Sapir et al (18), com-bined positron emission tomography (PET)/CT played a vital role in detecting recurrent disease and distinguishing it from postsurgical fibrosis. PET/CT allowed differentiation of benign from cancerous lesions with a sensitivity of 100% and a specificity of 96% (Fig 6) (18). Although the role of magnetic resonance (MR) imaging in the local staging of rectal cancer is well known, its role after surgical resection is less well defined. MR imaging has been shown to be a reliable technique for diagnosing pelvic recurrence of colorectal cancer but is not recommended for widespread routine surveillance (19,20).

Anterior ResectionAnterior resection, also referred to as abdomi-nal proctosigmoidectomy, involves resection of the rectosigmoid without perineal dissection, and subsequent anastomosis of the descend-ing colon and proximal rectum. Indications for anterior resection include sigmoid diverticular disease and rectal carcinoma more than 8 cm above the pectinate line (1). Because the anal sphincter is preserved, patients can expect a bet-ter functional outcome. There are three major types of anterior resection. High anterior resection involves excision of the distal descending colon, sigmoid colon, and upper rectum above the peritoneal reflection; standard low anterior resec-tion involves excision of the rectosigmoid and the upper and mid rectum below the peritoneal reflection; and extended low anterior resection in-volves excision of the distal sigmoid colon and the distal rectum (8). In the treatment of rectal cancer, standard low anterior resection is com-monly accompanied by total mesorectal resec-tion, with resection of the entire mesentery of the mid to lower rectum enveloped within the pelvic fascia containing the lymphatic drainage of the rectal tumor (11). This surgery has been associated with reduced local recurrence of


Figure 8. (a) CT image obtained in a 69-year-old man who had undergone anterior resection for rectal cancer shows the expected presacral fluid collection in the right posterior perirectal space (arrow). (b) Fol-low-up CT image obtained 7 months later shows a decrease in the amount of presacral fluid (arrow).

Figure 7. Axial (a) and sagittal reformatted (b) multidetector CT images obtained in a patient who had undergone low anterior resection show the low colorectal anastomotic staple line (white arrow), an increased amount of presacral fat (black arrow in a), and a posterior location of the seminal vesicles (arrowhead in a).

cancer of the mid to upper rectum (19). Bowel integrity is then restored, typically with creation of a conventional end-to-end coloanal anasto-mosis (Fig 7), although an end-to-side coloanal anastomosis and even a colonic pouch can be created. Newer surgical techniques that involve fashioning a colonic reservoir have increasingly supplanted the conventional approach. These colonic pouches have been shown to be associ-ated with a reduced incidence of anastomotic complications and better functional outcomes in comparison with the traditional coloanal anasto-mosis, and may be less likely to develop locally

recurrent tumor (9,21,22). With the creation of pouches, however, CT evaluation becomes more difficult due to the complex altered anatomy. In such cases in particular, knowledge of what sur-gery was performed is imperative to avoid misin-terpretation, and a discussion with the surgeon may be warranted.

The imaging findings following anterior resec-tion typically include a midline collection of fluid or a small amount of soft tissue in the presacral space (Figs 8, 9). The suture line is typically identifiable, and there can be wall thickening from postoperative edema at the anastomosis. The presacral space is usually larger, with the rectum displaced up to 2 cm anteriorly from the


Figure 11. Axial (a) and sagittal (b) CT images obtained in a patient with recurrent tumor who had un-dergone anterior resection show extensive infiltrative presacral soft tissue (white arrow) encasing the rectum (black arrow in a) above the anastomosis. Note the dilated loops of small bowel (black arrow in b), a finding that caused concern for associated obstruction secondary to the bulky tumor.

Figure 10. Sagittal reformatted multidetector CT image obtained in an 84-year-old man who had un-dergone anterior resection for proximal rectal cancer shows a slightly higher-than-expected location of the anastomosis (white arrow), with the expected increase in the size of the presacral space (black arrow).

Figure 9. CT image obtained in a patient who had undergone anterior resection shows a normal small amount of amorphous soft tissue in the presacral space (black arrow). Mildly prominent bowel loops suggestive of ileus (white arrow) are also seen.

sacrum (Fig 10) (1). More marked displacement (3–5 cm) or an increased amount of soft tissue should raise suspicion for an anastomotic leak or recurrent tumor (Fig 11). Extraperitoneal air or fluid around the iliac vessels has been described at CT and is more likely a benign finding that typically will resolve over a period of months (9). However, a similar finding has been reported in many patients with subclinical leaks after anterior resection and can persist for 6 months or more following surgery. The amount of air can also increase over time if there is increasing bowel lu-men distention and pressure (9).

Hartmann ProcedureThe Hartmann procedure involves fashioning a temporary diverting colostomy after distal colec-tomy or sigmoid colon resection, with a sutured rectal or colonic stump left behind for a possible future elective reanastomosis. In contrast, a mu-cous fistula is created when the stump is brought up to the abdominal wall, although doing so is not always possible owing to the length of the stump. A surgeon may create a mucous fistula when a second anastomotic procedure is contemplated,


Figure 14. CT image shows a long, edematous Hart-mann pouch with mild diverticulitis (white arrow) and, at its most proximal aspect, a focal mucocele (black ar-row), findings that were confirmed with endoscopy.

Figure 13. CT image shows the normal appearance of a Hartmann pouch (white arrow), which may con-tain residual fluid, air, or debris. The suture (black ar-row) marks the blind end.

since it is typically easier to find. In the Hartmann procedure, the sutured stump varies in length and may be as distal as the rectum and as proximal as the transverse colon (21). The Hartmann pro-cedure is often performed in patients who are at high risk for anastomotic leak or subsequent complications of peritonitis and sepsis following an immediate primary anastomosis (1). The pro-cedure is commonly performed in emergent cases of complicated acute diverticulitis, sigmoid per-foration, or obstruction secondary to a carcinoma (21). The diverting colostomy allows time for the bowel to heal before a “takedown” is performed in the second stage of the surgery to restore lumi-nal continuity.

Historically, primary evaluation of the colonic or rectal stump was performed with water-based contrast material–enhanced radiography to assess the integrity of the pouch. Because rupture of the Hartmann pouch is associated with considerable morbidity and mortality, a preliminary pouch study is typically performed after the first stage of the surgical procedure (21) and before surgical planning of the colostomy takedown. Multide-tector CT also clearly depicts the location of the temporary diverting colostomy or ileostomy and of the rectal or colonic stump, which is typically closed with metallic staples or sutures at its blind end (Fig 12). The pouch may occasionally be left open, particularly in difficult surgical cases, with a drain left within the pouch. In these cases, if an abscess subsequently develops, it can be drained either through the indwelling drain in the pouch or transgluteally. The length of the pouch and its relation to the colostomy are also important to consider and can be depicted with multidetector

Figure 12. CT image obtained in a 63-year-old man with a history of Crohn disease and severe diverticulitis who had undergone creation of a Hartmann pouch and a temporary diverting ileostomy (not shown) 1 month earlier. The Hartmann pouch demonstrates mild wall thickening (arrow).

CT. It has been reported that the length of the pouch decreases over time, so that its original anatomic position can be altered. Occasionally, the pouch can be sutured to the abdominal wall or brought up to the abdominal wall as a mu-cous fistula. At multidetector CT, the pouch may contain polyps, diverticula, debris, or inspissated mucus (Figs 13, 14). In addition to radiography or CT, endoscopy is often used to evaluate the pouch prior to surgery. At this time, diversion (or diversional) colitis can be a common and even expected endoscopic finding, typically resolving later with reanastomosis (discussed later).

Restorative Proctocolectomy with IPAARestorative proctocolectomy is the definitive sur-gical treatment in patients with ulcerative colitis, familial adenomatous polyposis, multiple colon cancers, or Lynch syndrome (1). The procedure involves a total proctocolectomy and creation


Figure 15. Axial (a) and sagittal (b) multidetector CT images obtained in a 48-year-old woman who had undergone restorative proctocolectomy with IPAA for ulcerative colitis show the configuration of the J pouch. Note the linear row of staples (white arrow in a), the terminal blind-end staple line (black arrow in a), and the anastomosis between the ileal pouch and the anus (arrow in b).

of an ileal pouch and an IPAA. Common con-traindications for the procedure include Crohn disease (due to the possibility of recurrence in the ileal pouch), significant anorectal disease, and preexisting incontinence (23). Restorative proctocolectomy consists of an ileoanal anasto-mosis with an ileal reservoir fashioned from the terminal ileum. There are three main types of ileal pouches—J, S, and W pouches—whose names refer to the shape of the reservoir (24). The most commonly created pouch is the J pouch, since its creation involves a relatively less complex tech-nique. The ileal J pouch is created by folding 15 cm of terminal ileum and fashioning a side-to-side anastomosis with a linear stapler. The anas-tomosis to the anus is created along the pectinate line (25). Newer techniques may leave more anoderm in place, as well as a transitional zone of epithelium. These pouches can then be com-plicated with recurrent active ulcerative colitis and resemble postoperative pouchitis. The newly created pouch is allowed to heal by fashioning a proximal temporary loop ileostomy, which is taken down in the second stage of the procedure after approximately 8–12 weeks (26). Sometimes the procedure consists of a single stage, without the creation of an ileostomy, thereby prevent-ing the possible complications associated with a loop ileostomy. A disadvantage is the associated increased risk of pelvic and peritoneal sepsis from a pouch or anal anastomotic leak (11). Although the creation of an ileoanal pouch anastomosis is associated with a low mortality rate of less than 1%, a morbidity rate ranging from 18% to 70% has been reported (26). In patients with ulcer-ative colitis, approximately 20% of pouches have to be removed owing to complications (27).

Typically, fluoroscopic pouchography has rou-tinely been performed after IPAA before closure of the ileostomy. Rectal administration of a water-soluble contrast agent allows evaluation of the integrity of the ileal pouch and IPAA before take-down of the temporary ileostomy. The distended pouch should be visualized in different imaging planes. In addition, postevacuation views should be obtained to detect subtle leaks or strictures that could have been obscured (25). However, multi-detector CT is currently the favored modality for postoperative assessment of the IPAA in symptom-atic patients with clinical suspicion for infection or other complications. Intravenous contrast mate-rial is used, as well as retrograde administration of water-soluble contrast material to adequately distend the pouch. A typically fluid-filled structure with a row of metallic staples in the blind-ending ileum and a double row of staples at the ileoileal anastomosis helps identify the pouch (Figs 15, 16) (24). The ileoanal anastomosis is character-ized by radiodense staples oriented 180° from each other (28). The pouch has clear margins and is located close to the sacrum posteriorly and the bladder anteriorly (29). Following takedown of the temporary diverting loop ileostomy, oral contrast-enhanced multidetector CT can be performed to exclude any extraluminal fluid, contrast material, or air (28). Mild stranding in the peripouch fat is another finding that is normally seen in the im-mediate postoperative period. This finding may be related to preexisting chronic inflammation from long-standing ulcerative colitis and does not nec-essarily indicate acute infection (25). Post-IPAA pouch bleeding has been reported in a minority


Figure 17. Multidetector CT image obtained in a patient who had undergone a right hemicolectomy for a colonic mass demonstrates an ileocolic end-to-side anastomosis (arrow).

Figure 16. Coronal reformatted multidetector CT image obtained in a 59-year-old man who had under-gone restorative proctocolectomy with IPAA for ulcer-ative colitis demonstrates the ileoileal vertical row of staples (black arrow). Note that the pouch is filled with contrast material (white arrow).

of patients and is usually managed nonoperatively with clot evacuation and other noninvasive tech-niques of hemostasis.

Segmental ResectionsSegmental resections of the colon include ileoce-cal, right, transverse, and left colectomies and are based on the location of the pathologic condition. Crohn ileitis is a common indication for ileocecal resection (2). Indications for right and transverse colectomies include carcinoma and complications of IBD. Left colectomies are typically performed for left-sided tumors or complicated diverticular disease (8). Segmental resections for malignant neoplasms require wide margins with an ex-tended segmental colectomy, as well as mesen-teric and omental resections with excision of the lymphatic drainage of the tumor.

Following these segmental resections, a straight anastomosis is usually fashioned. There are three types of anastomoses. The end-to-end type is indicated when the bowel lumina are of roughly equal circumference (eg, in rectal resection). The end-to-side type is used when the proximal end is a smaller lumen that is con-nected to the side of the larger lumen (eg, in an ileocolic anastomosis after a right hemicolec-tomy) (Fig 17). The side-to-side type involves the connection of two antimesenteric borders of the bowel. It has the advantage of forming a large, well-vascularized anastomosis and is commonly used in ileocolic and small bowel anastomoses (8). A side-to-side stapled anastomosis is usu-ally avoided on the left side, since it can make subsequent colonoscopy difficult. Multidetec-tor CT findings following segmental resection may include absence of excised bowel segments, anastomotic surgical clips, and displacement of the adjacent viscera into the unoccupied postop-erative spaces (1).

Common Postsurgical Complications

Postsurgical complications are not uncommon after bowel resection surgery. The complications vary according to (a) the surgical procedure that was performed, (b) the indication for the surgery, (c) the length of the postoperative period (1), and (d) the presence of any underlying concomitant disease entities (2). As mentioned earlier, com-mon specific postoperative complications include wound complications, anastomotic leak, fistula, intraperitoneal abscess, pouchitis, small bowel

obstruction and pseudo-obstruction, and diver-sion colitis. Multidetector CT with the use of multiplanar reformation provides good anatomic detail and allows early evaluation of the location, extent, and type of postsurgical complications after colon resection (3–10).


Figure 18. CT image obtained in a patient with wound dehiscence demonstrates a fluid collection within the abdominal wall defect (arrow) and separa-tion of the abdominal wall layers.

Figure 19. CT image demonstrates apparent wound dehiscence (arrows). Rather, the wound was left to heal by secondary intention after surgery and was initially misinterpreted as wound dehiscence.

Figure 20. CT image obtained in a patient who had undergone APR shows a colostomy (arrowhead) and a stomal hernia containing only fat and mesenteric ves-sels (arrow).

Wound ComplicationsAbout 5%–10% of patients who have undergone major abdominal surgery develop wound infec-tions (4). Despite adequate bowel preparation and prophylactic antibiotics, the risk for infection remains high and is often due to contamination by anaerobic microbes. This can lead to inflam-mation, infection, and necrosis at or adjacent to the wound site (4). Wound dehiscence, abdomi-nal incisional hernia, peritonitis, and systemic sepsis are potential sequelae (30–32). Perineal resections in APRs are specifically associated

with varying degrees of wound complications, from minor wound dehiscence to fistula forma-tion, sinuses, and perineal hernias. Risk factors associated with wound complications include an increased patient body mass index and preexist-ing chronic medical conditions, including IBD, malnutrition, and steroid use (13). Treatment op-tions include administration of broad-spectrum antibiotics or incision and drainage of potentially infected fluid collections (30–32).

Early CT findings of wound dehiscence include accumulations of fluid with air within the wound and adjacent tissues (Fig 18). Careful observation of the abdominal wall layers around the incision site―including the skin, subcutaneous tissue, linea alba, anterior and posterior fascial sheaths, extraperitoneal fat, and peritoneum―is important to evaluate for separation between the layers of tissue (33). Fascial separation in wound dehis-cence should be distinguished from overt wound evisceration, which represents an emergent surgi-cal indication. Secondary intention wound closure and wound packing with surgical dressing may also mimic wound dehiscence (Fig 19).

Incisional hernia is more common with verti-cal than with horizontal incisions, with a reported prevalence of 0.5%–50% in abdominal surgery patients (30). A stomal or parastomal hernia is a common form of incisional hernia of bowel or mesenteric fat that occurs at the site of or adja-cent to the stoma (Fig 20). It may be an expected


Figure 21. CT image obtained in a 56-year-old woman who had undergone low anterior resection demonstrates the double rectum sign. Note the extrav-asation of contrast material and air (black arrow) ad-jacent to and mimicking the true rectum (white arrow).

finding after stoma creation and may not cause the surgeon concern unless it is complicated by bowel obstruction or perforation. Fat stranding, fluid accumulation within the stoma site, and thickening of the involved bowel wall are addi-tional findings that suggest early bowel strangula-tion from incarceration.

Anastomotic LeakAn anastomotic leak after bowel surgery is as-sociated with high morbidity and mortality and may result in complications, including peritonitis and sepsis (34,35). The treatment of leaks often requires further invasive procedures, including percutaneous drain creation, diverting ostomies, and repeat open surgery. The mortality rate can be as high as 50% (3) if anastomotic leaks are not identified and treated appropriately. Leaks have been reported to account for one-third of deaths after colorectal surgery, with a reported incidence ranging from 2% to 51% (36). No significant dif-ference in the incidence of leaks has been reported between “hand-sewn” and “stapled” surgical tech-niques (37). Risk factors have been implicated in the development of leaks, including malnutrition, male gender, smoking, and steroid use. Periopera-tive factors such as intraoperative sepsis, perioper-ative bleeding, and excessive anastomotic tension also increase the risk for an anastomotic leak (38). A low anterior rectal anastomosis carries the high-est risk for developing an anastomotic leak (35). In IPAA, anastomotic leaks most frequently occur at the pouch-anal anastomosis but can also occur at the ileoileal staple line of the pouch (pouch dehis-cence) or from the pouch itself (29).

Clinical symptoms suggestive of an anastomotic leak usually manifest in the first 2 postoperative weeks, most commonly between days 5 and 7. Symptoms include fever, intense abdominal pain, tachycardia, guarding, and rebound tenderness (4). Tenesmus or constipation may be present, al-though these are more nonspecific symptoms that can also occur with tumor recurrence. Oftentimes, early clinical manifestations may be subtle and nonspecific and therefore difficult to recognize. Leak incidence has been reported to range from 3% to 40%, and up to as high as 69% in another study in which leaks were diagnosed with barium enema examination (39,40). It is reported that subclinical leaks have been discovered at routine postoperative radiologic examinations in 5.7%–10.7% of cases (39). These occult leaks can persist for years without significant symptoms. They are commonly seen following procedures in which a temporary loop ileostomy is created to protect the distal anastomosis (eg, IPAA). Although ileostomy creation may decrease the risk of associated com-plications of anastomotic leak, it does not appear

to decrease the overall incidence of leak (37). Sub-clinical leaks do appear to have a higher incidence of spontaneous healing compared with clinical leaks (41).

Although fluoroscopic imaging of the bowel with rectal administration of water-based contrast material has been shown to be useful in identifying the exact location and features of an anastomotic leak (4), multidetector CT is currently the imag-ing modality of choice to evaluate for suspected anastomotic leak and has been reported to confirm clinical leakage in 48%–100% of cases (9,36). Multidetector CT should be performed with rectal contrast material, since oral contrast material may not have had time to reach the area of the leak. Ad-ministration of rectal contrast material immediately prior to multidetector CT and close evaluation of the multiplanar reformatted images will better define a presacral collection for potential areas of leak. An anastomotic leak can be diagnosed with confidence at multidetector CT when there is frank extravasation of bowel contrast material with air or possibly stool. After anterior resection, this ap-pearance can mimic the true rectum and has been described as the “double rectum” sign (Fig 21) (9). Less definitive but suggestive signs include extralu-minal air in a higher proportion relative to fluid in the newly formed postoperative spaces. Accumula-tion of air in the presacral space beyond 6 months after surgery is suggestive of a leak (9). Perianasto-motic loculated fluid containing air or dispropor-tionate fluid and air has been reported to be the most common indicator of an anastomotic leak (Fig 22) (42). These fluid and gas collections can also be associated with anterior displacement of the rectum of more than 5 cm from the sacrum (1).

It should be noted that radiologic findings consistent with an anastomotic leak do not always determine the treatment approach or change the clinical management. Rather, management is often based on clinical findings. Leaks do not always re-


quire surgical management because many of them, including occult leaks, appear to be self limiting. If the leak is contained and the patient is stable with-out signs of sepsis or peritonitis, initial treatment may be conservative and consist of the administra-tion of antibiotics (39). Multidetector CT can be of use in guiding percutaneous drainage.

Various studies have reported false-negative rates of 35%–49% for radiologic imaging of suspected anastomotic leaks (27,40,43). A false-negative multidetector CT diagnosis for a leak is typically related to the timing of the postop-erative radiographic study or to the factors that influence the quality of the study, such as dis-tention of the anastomotic segment, use of rectal contrast agent, and use of thin axial sections and multiplanar reformation. The false-negative rate may be higher in the immediate postoperative period, before the leaked fluid and air have had time to accumulate. A distal anastomotic leak may be missed if the rectal contrast material catheter seals it off, particularly if the anasto-mosis is low and a balloon is used as opposed to a straight catheter (9). The best way to define a

low anastomotic leak is at CT performed with oral, intravenous, and rectal contrast material. Occasionally, the rectum may collapse during CT due to intermittent spasm, which can limit detection of a leak or lead to underestimation of the size of the leak.

FistulaOngoing leaks may result in fistulas or abscesses. Fistulas can involve the ileal or Hartmann pouch, anus, vagina, or bladder or extend to the skin (enterocutaneous fistula). After ileoanal recon-struction, fistulas may occur secondary to a leak or develop as a late complication. Multidetector CT is useful in defining perianal and perineal fis-tulas, which are often complex. Fistulas involving the surgically created pouches can be seen as en-hancing tracts of contrast material extending into the peripouch soft tissues with associated sur-rounding fat stranding (Fig 23). Simple fistulas can be treated with resection or seton placement. Abscess drainage, diversion, and antibiotics may help bring about spontaneous closure. Recurrent fistulas may require a permanent stoma (22).

Figure 22. CT image obtained in an 85-year-old man who had undergone an-terior resection for rectal adenocarcinoma clearly demonstrates dehiscence of the su-ture line (black arrow), with a collection of contrast material and air in the left perirectal region (white arrow).

Figure 23. Consecutive CT images obtained in a patient with recurrent IBD demonstrate a rim-enhancing tract (arrows in a) containing fluid and air and extending from the IPAA–J pouch (arrows in b) to the anterior abdominal wall, findings that are consistent with an enterocutaneous fistula.


Figure 26. CT image shows Hartmann pouch dehis-cence (arrowhead) with pelvic abscesses (black arrow) and enhancing peritoneum (white arrow).

Figure 24. CT image obtained in a 48-year-old man who had undergone APR for recurrent anal squamous cell carcinoma shows a rim-enhancing posterior pel-vic fluid collection (arrow), a finding that is consistent with abscess.

Figure 25. CT image shows a right pelvic abscess (white arrow) secondary to an IPAA pouch leak (black arrow).

Intraperitoneal AbscessIntraabdominal abscesses remain the most com-mon cause of morbidity following colorectal surgery (1) and have a high potential for serious ramifications when associated with systemic sep-sis. The morbidity and mortality rate has been reported as 10%–40% (4). Common associations include anastomotic leaks, wound infections, fistulas, and contamination during surgery. The usual causative agents are enteric microbes such as Clostridium, Proteus, Escherichia coli, Klebsiella, and Bacteroides (4). Multidetector CT has been reported to be sensitive and specific—more sen-sitive than pouchography or fluoroscopy—with an accuracy of 90% in the evaluation of abscess size and location (4). Multidetector CT findings include a well-defined and well-circumscribed, low-attenuation spherical or ellipsoid collection of fluid. There is usually a peripheral enhancing rim, and there can be septations, gas, or gas-fluid levels in the collection (Figs 24, 25). Associated thicken-ing, enhancement, and obliteration of adjacent fascial and peritoneal planes are well delineated (Fig 26). Characterization of the size and location of the abscess at CT will guide either surgical or percutaneous management (4). In a Hartmann pouch, dehiscence of a short stump can result in extraperitoneal leakage and subsequent pelvic ab-scess formation. In contrast, dehiscence in a longer

pouch can lead to intraperitoneal leakage, intraab-dominal abscesses, and peritonitis, often necessi-tating immediate surgical repair.

PouchitisPouchitis, as its name indicates, is nonspecific inflammation of a pouch, and it may be acute or chronic. Pouchitis is usually an early postopera-tive complication and is the most common indi-cator of pouch dysfunction (26). It is extremely common after restorative proctocolectomy with IPAA, affecting up to 50% of patients (44–46). In patients with prior ulcerative colitis, the incidence of pouchitis ranges from 20% to 50%, compared with only 0%–10% in patients treated for familial


Figure 28. CT image obtained in a 48-year-old pa-tient who had undergone IPAA 8 weeks earlier demon-strates mild wall thickening of the rectal pouch (arrow) with a moderate amount of surrounding fluid. Endos-copy showed mild erythematous changes consistent with early pouchitis.

Figure 27. Coronal reformatted multidetector CT image demonstrates wall thickening and enhancement (white arrow) with fluid distention of a rectal pouch (black arrow), as well as perirectal stranding and infil-tration, findings that are consistent with pouchitis.

adenomatous polyposis (44). The specific cause of pouchitis is currently unknown, but dysbiosis (changes in microflora in the pouch lumen), fe-cal stasis within the pouch, and ischemia have been described as possible causes (44). Patients at increased risk for the development of pouchitis include those with severe ulcerative colitis or ex-traintestinal manifestations of IBD and those who chronically use nonsteroidal anti-inflammatory drugs (46). Clinical manifestations of pouchitis include increased stool frequency, bright red rec-tal bleeding, fever, malaise, crampy abdominal pain, diarrhea, and hematochezia (29). However, these clinical signs are nonspecific and can also suggest anastomotic leak or abscess, so that CT is often indicated. A sensitivity of 71% has been re-ported for CT in the detection of pouchitis (28); however, the features are also nonspecific and include bowel wall thickening, pouch dilatation, peripouch adenopathy, fat stranding, fluid ac-cumulation, and mucosal enhancement (Fig 27) (23). Mild wall thickening may be obscured by poor distention. Evaluation for pouchitis should not be performed until at least 6–8 weeks after surgery to allow resolution of the normal postop-erative changes and edema (Fig 28). The actual diagnosis is made with endoscopy and biopsy (25,26). Pouchography with fluoroscopy is usu-ally reserved for evaluating pouch function (defe-cography and pouch distention).

Small Bowel Obstruc- tion and Pseudo-obstructionSmall bowel obstruction is a common postop-erative complication. It is often seen after IPAA (15%–44% of cases) and typically occurs at least 30 days after takedown (28,34). Because the distal ileum is used to make the pouch, when it is pulled into the pelvis, it stretches the mesentery and can cause the proximal small bowel to herniate and be-come obstructed. Other important causes of post-operative bowel obstruction include adhesions, strictures, and volvulus. Early postoperative small bowel obstruction is treated conservatively with nasogastric tube decompression, whereas later cases often require surgery (29). Multidetector CT of bowel obstruction typically demonstrates dis-tended bowel loops proximal to a transition point. Tumor recurrence or any sign of an inflammatory process should be excluded. Rarely, twisting of the mesenteric vessels is seen at the transition point, a finding that indicates a volvulus (47). Coronal re-formatted images have been shown to be valuable in identifying the transition point (39). Anal stric-tures from fibrosis or anastomotic ischemia are not uncommon after IPAA (34).

In contrast, postoperative ileus (pseudo-ob-struction) is a common cause of bowel dilatation in the immediate postoperative period following colon surgery (4). Clinically, patients will appear to have a colonic obstruction, although imaging


Figure 30. CT image obtained in a patient with an ileostomy and diversion colitis shows wall thickening, stratification, and fat stranding around the right (arrow-head), transverse (white arrow), and descending (black arrow) colon which is diverted from the fecal stream.

Figure 29. CT image of the ileus in a patient who had undergone colonic resection shows diffusely and mildly dilated loops of small bowel (black arrow) and colon (white arrow) with air-fluid levels and no identi-fiable obstructive lesion.

will fail to show any identifiable physical cause of obstruction (Fig 29). The condition infrequently lasts longer than 48 hours; therefore, persistence of symptoms 2–5 days after surgery should raise suspicion for other causes (48,49). Persistent clinical symptoms after conservative manage-ment may require evaluation with multidetector CT (10). Multidetector CT is often the initial test after abdominal radiography for small bowel obstruction and has replaced traditional barium enema examination. A common multidetector CT finding is dilatation of the ascending and transverse colon proximal to the splenic flexure and relative distal colonic decompression without identification of an obstructing lesion (49).

Diversion ColitisDiversion colitis is a nonspecific mucosal inflam-mation of any colonic or rectal segment that has been diverted from the fecal stream in a loop or end colostomy (50). The condition is seen to some extent in nearly all patients with diversion and is asymptomatic in the majority of cases. Symptoms (if present) may include abdomi-nal pain, tenesmus, hematochezia, and mucous discharge (51). Symptoms and inflammation promptly subside after reestablishment of bowel continuity, which is the definitive treatment. The cause has not been well defined but has been proposed to be deprivation of the colonic muco-sal cells of the diverted segment of essential nu-trients and by-products from enteric bacteria in

stool (including short-chain fatty acids) (51,52). Alternatively, the inflammation may be caused by stasis within the diverted segment (52). There is little in the literature about the radiologic features of diversion colitis. Features at double-contrast barium enema examination have been reported to include diffuse mucosal nodularity, inflamma-tory pseudopolyps, punctate ulcerations, and (in more chronic cases) a diffuse granular mucosal appearance (52). Multidetector CT demonstrates features of a nonspecific colitis (Fig 30), and the involved mucosa may mimic ulcerative colitis (Fig 31). Pouchitis, which occurs in continent pouches such as Kock pouches or IPAAs, should be distinguished from diversion colitis, which occurs in blind pouches. Long-standing chronic

Figure 31. CT image shows diversion colitis in a Hartmann pouch, which is thick walled and distended by fluid (arrow).


inflammation in diversion colitis may increase the risk of developing malignancy (53). Oftentimes, negative cultures can be used to rule out a super-imposed bacterial or parasitic infection. Absence of a history of recent antibiotic use and lack of characteristic pseudomembranes at endoscopy exclude a possible C difficile infection. Therefore, diversion colitis should be considered when post-operative nonspecific proctocolitis is seen in an otherwise asymptomatic patient with a colostomy. The patient should then be evaluated for treat-ment with reanastomosis (51).

ConclusionMultidetector CT plays an integral role in the evaluation of the postoperative colon. The normal postoperative anatomy and common postoperative complications are well demonstrated with axial multidetector CT supplemented with multiplanar reformation, which can help distinguish benign expected findings from those associated with more worrisome disease entities. When possible, the radiologist should communicate with the surgeon to find out specifically what surgery was per-formed, what anastomosis was created, and what the indication is for the CT examination. Accurate characterization will facilitate prompt diagnosis of potentially life-threatening complications and aid the surgeon with clinical management.

Acknowledgments.—We acknowledge Michael A. Wein-stein, MD, and Anthony M. Sajewicz, MD, for their contributions to the manuscript, and Donald Chau for his administrative support.

References 1. Zissin R, Gayer G. Postoperative anatomic and

pathologic findings at CT following colonic re-section. Semin Ultrasound CT MR 2004;25(3): 222–238.

2. Scardapane A, Brindicci D, Fracella MR, Angelelli G. Post colon surgery complications: imaging find-ings. Eur J Radiol 2005;53(3):397–409.

3. Guillem JG, Cohen AM. Current issues in colorectal cancer surgery. Semin Oncol 1999;26 (5):505–513.

4. Gore RM, Berlin JW, Yaghmai V, Mehta U, New-mark GM, Ghahremani GG. CT diagnosis of postoperative abdominal complications. Semin Ul-trasound CT MR 2004;25(3):207–221.

5. Ghahremani GG, Gore RM. CT diagnosis of post-operative abdominal complications. Radiol Clin North Am 1989;27(4):787–804.

6. Reznek RH, White FE, Young JW, Fry IK, Nicholls RJ. The appearances on computed tomography after abdominoperineal resection for carcinoma of the rectum: a comparison between the normal appearances and those of recurrence. Br J Radiol 1983;56(664):237–240.

7. Horton KM, Abrams RA, Fishman EK. Spiral CT of colon cancer: imaging features and role in man-agement. RadioGraphics 2000;20(2):419–430.

8. Dunn KMB, Rothenberger DA. Colon, rectum and anus. In: Brunicardi FC, ed. Schwartz’s prin-ciples of surgery. 9th ed. New York, NY: McGraw Hill, 2010; 1013–1072.

9. DuBrow RA, David CL, Curley SA. Anastomotic leaks after low anterior resection for rectal carci-noma: evaluation with CT and barium enema. AJR Am J Roentgenol 1995;165(3):567–571.

10. Kelvin FM, Korobkin M, Heaston DK, Grant JP, Akwari O. The pelvis after surgery for rectal carcinoma: serial CT observations with emphasis on nonneoplastic features. AJR Am J Roentgenol 1983;141(5):959–964.

11. Fry RD, Mahmoud N, Maron DJ, et al. Colon and rectum. In: Townsend CM, Beauchamp DR, Evers M, Mattox KL, eds. Sabiston’s textbook of surgery. 18th ed. Philadelphia, Pa: Saunders, 2008; 1348–1432.

12. Murrell ZA, Dixon MR, Vargas H, Arnell TD, Ku-mar R, Stamos MJ. Contemporary indications for and early outcomes of abdominoperineal resection. Am Surg 2005;71(10):837–840.

13. Ogilvie JW, Ricciardi R. Complications of perineal surgery. Clin Colon Rectal Surg 2009;22(1):51–59.

14. Mulsow J, Winter DC. Sphincter preservation for distal rectal cancer: a goal worth achieving at all costs? World J Gastroenterol 2011;17(7):855–861.

15. Lee JK, Stanley RJ, Sagel SS, Levitt RG, McClen-nan BL. CT appearance of the pelvis after abdom-ino-perineal resection for rectal carcinoma. Radiol-ogy 1981;141(3):737–741.

16. Husband JE, Hodson NJ, Parsons CA. The use of computed tomography in recurrent rectal tumors. Radiology 1980;134(3):677–682.

17. Moss AA, Thoeni RF, Schnyder P, Margulis AR. Value of computed tomography in the detection and staging of recurrent rectal carcinomas. J Com-put Assist Tomogr 1981;5(6):870–874.

18. Even-Sapir E, Parag Y, Lerman H, et al. Detection of recurrence in patients with rectal cancer: PET/CT after abdominoperineal or anterior resection. Radiology 2004;232(3):815–822.

19. Goldberg S, Klas JV. Total mesorectal excision in the treatment of rectal cancer: a view from the USA. Semin Surg Oncol 1998;15(2):87–90.

20. Titu LV, Nicholson AA, Hartley JE, Breen DJ, Monson JR. Routine follow-up by magnetic reso-nance imaging does not improve detection of re-sectable local recurrences from colorectal cancer. Ann Surg 2006;243(3):348–352.

21. Cherukuri R, Levine MS, Maki DD, Rubesin SE, Laufer I, Rosato EF. Hartmann’s pouch: radio-graphic evaluation of postoperative findings. AJR Am J Roentgenol 1998;171(6):1577–1582.

22. Dehni N, Parc R, Church JM. Colonic J-pouch–anal anastomosis for rectal cancer. Dis Colon Rec-tum 2003;46(5):667–675.

23. Broder JC, Tkacz JN, Anderson SW, Soto JA, Gupta A. Ileal pouch–anal anastomosis surgery: imaging and intervention for postoperative compli-cations. RadioGraphics 2010;30(1):221–233.

24. Pizzamiglio M, Catalano C, Sarrantonio A, et al. Restorative proctocolectomy: morphological and functional study with coronal CT. Abdom Imaging 1996;21(3):261–265.

25. Crema MD, Richarme D, Azizi L, Hoeffel CC, Tu-biana JM, Arrivé L. Pouchography, CT, and MRI


features of ileal J pouch–anal anastomosis. AJR Am J Roentgenol 2006;187(6):W594–W603.

26. Beliard A, Prudhomme M. Ileal reservoir with ileo-anal anastomosis: long-term complications. J Visc Surg 2010;147(3):e137–e144.

27. Akyol AM, McGregor JR, Galloway DJ, George WD. Early postoperative contrast radiology in the assessment of colorectal anastomotic integrity. Int J Colorectal Dis 1992;7(3):141–143.

28. Seggerman RE, Chen MY, Waters GS, Ott DJ. Radiology of ileal pouch–anal anastomosis surgery. AJR Am J Roentgenol 2003;180(4):999–1002.

29. Alfisher MM, Scholz FJ, Roberts PL, Counihan T. Radiology of ileal pouch–anal anastomosis: normal findings, examination pitfalls, and complications. RadioGraphics 1997;17(1):81–98; discussion 98–99.

30. Ghahremani GG, Jimenez MA, Rosenfeld M, Rochester D. CT diagnosis of occult incisional her-nias. AJR Am J Roentgenol 1987;148(1):139–142.

31. Marn CS. Anterior abdominal wall. In: Gore RM, Levine MS, eds. Textbook of gastrointestinal radi-ology. 2nd ed. Philadelphia, Pa: Saunders, 2000; 2010–2019.

32. Rout WR. Complications of incision. In: Zuidema GD, ed. Surgery of the alimentary tract. 4th ed. Philadelphia, Pa: Saunders, 1996; 332–335.

33. Smith-Behn J, Arnold M, Might J. Use of comput-erized tomography of the abdominal wall in the di-agnosis of partial post-operative wound dehiscence. Postgrad Med J 1986;62(732):947–948.

34. Gorgun E, Remzi FH. Complications of ileoanal pouches. Clin Colon Rectal Surg 2004;17(1):43–55.

35. Bertelsen CA, Andreasen AH, Jørgensen T, Harling H; Danish Colorectal Cancer Group. Anastomotic leakage after anterior resection for rectal cancer: risk factors. Colorectal Dis 2010;12(1):37–43.

36. Nicksa GA, Dring RV, Johnson KH, Sardella WV, Vignati PV, Cohen JL. Anastomotic leaks: what is the best diagnostic imaging study? Dis Colon Rec-tum 2007;50(2):197–203.

37. Kumar A, Daga R, Vijayaragavan P, et al. Anterior resection for rectal carcinoma: risk factors for anas-tomotic leaks and strictures. World J Gastroenterol 2011;17(11):1475–1479.

38. Chambers WM, Mortensen NJ. Postoperative leak-age and abscess formation after colorectal surgery. Best Pract Res Clin Gastroenterol 2004;18(5): 865–880.

39. Bielecki K, Gajda A. The causes and prevention of anastomotic leak after colorectal surgery [in Czech]. Klin Onkol 1999;12:25–30. http://www .linkos.cz/files/klinicka-onkologie/59/1379.pdf.

40. Goligher JC, Graham NG, De Dombal FT. Anas-tomotic dehiscence after anterior resection of rec-tum and sigmoid. Br J Surg 1970;57(2):109–118.

41. Lim M, Akhtar S, Sasapu K, et al. Clinical and subclinical leaks after low colorectal anastomosis: a clinical and radiologic study. Dis Colon Rectum 2006;49(10):1611–1619.

42. Power N, Atri M, Ryan S, Haddad R, Smith A. CT assessment of anastomotic bowel leak. Clin Radiol 2007;62(1):37–42.

43. Haynes IG, Goldman M, Silverman SH, Alex-ander-Williams J, Keighley MR. Water-soluble contrast enema after colonic anastomosis. Lancet 1986;1(8482):675–676.

44. McLaughlin SD, Clark SK, Tekkis PP, Nicholls RJ, Ciclitira PJ. The bacterial pathogenesis and treat-ment of pouchitis. Therap Adv Gastroenterol 2010; 3(6):335–348.

45. Thoeni RF, Fell SC, Engelstad B, Schrock TB. Ileoanal pouches: comparison of CT, scintigraphy, and contrast enemas for diagnosing postsurgical complications. AJR Am J Roentgenol 1990;154(1): 73–78.

46. Wu H, Shen B. Pouchitis and pouch dysfunction. Med Clin North Am 2010;94(1):75–92.

47. Sinha R, Verma R. Multidetector row computed tomography in bowel obstruction. I. Small bowel obstruction. Clin Radiol 2005;60(10):1058–1067.

48. Sandrasegaran K, Maglinte DDT. Imaging of small bowel-related complications following major ab-dominal surgery. Eur J Radiol 2005;53(3):374–386.

49. Frager D, Rovno HD, Baer JW, Bashist B, Fried-man M. Prospective evaluation of colonic obstruc-tion with computed tomography. Abdom Imaging 1998;23(2):141–146.

50. Glotzer DJ, Glick ME, Goldman H. Proctitis and colitis following diversion of the fecal stream. Gas-troenterology 1981;80(3):438–441.

51. Whelan RL, Abramson D, Kim DS, Hashmi HF. Diversion colitis: a prospective study. Surg Endosc 1994;8(1):19–24.

52. Scott RL, Pinstein ML. Diversion colitis demon-strated by double-contrast barium enema. AJR Am J Roentgenol 1984;143(4):767–768.

53. Lafreniere R, Ketcham AS. Hartmann’s pouch car-cinoma. J Surg Oncol 1985;29(1):26–27.

This journal-based SA-CME activity has been approved for AMA PRA Category 1 CreditTM. See www.rsna.org/education/search/RG.

Teaching Points March-April Issue 2013

Multidetector CT of the Postoperative Colon: Review of Normal Appear-ances and Common ComplicationsStefanie Weinstein, MD • Samuel Osei-Bonsu, MD • Rizwan Aslam, MBChB • Judy Yee, MD

RadioGraphics 2013; 33:515–532 • Published online 10.1148/rg.332125723 • Content Codes:

Page 518Difficulty has been reported in distinguishing the characteristic finding of postoperative presacral soft tissue from tumor recurrence (15–17). However, evolution of these changes on serial CT images helps in differentiation (Fig 5).

Page 519At other times, the presacral soft tissue remains stable and can persist for up to 2 years after surgery, or even indefinitely.

Page 523However, multidetector CT is currently the favored modality for postoperative assessment of the IPAA in symptomatic patients with clinical suspicion for infection or other complications. Intravenous contrast ma-terial is used, as well as retrograde administration of water-soluble contrast material to adequately distend the pouch.

Page 526Perianastomotic loculated fluid containing air or disproportionate fluid and air has been reported to be the most common indicator of an anastomotic leak (Fig 22) (42).

Page 526It should be noted that radiologic findings consistent with an anastomotic leak do not always determine the treatment approach or change the clinical management. Rather, management is often based on clinical findings.

multidetector ct of the post- operative colon: review of normal

Documents