Livia Biancone and Emma Calabrese contributed equally to this work.
Wireless capsule endoscopy and small intestine contrast ultrasonography in recurrence of Crohn's disease
Article first published online: 18 JUN 2007
Copyright © 2007 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 13, Issue 10, pages 1256–1265, October 2007
How to Cite
Biancone, L., Calabrese, E., Petruzziello, C., Onali, S., Caruso, A., Palmieri, G., Sica, G. S. and Pallone, F. (2007), Wireless capsule endoscopy and small intestine contrast ultrasonography in recurrence of Crohn's disease. Inflamm Bowel Dis, 13: 1256–1265. doi: 10.1002/ibd.20199
- Issue published online: 11 SEP 2007
- Article first published online: 18 JUN 2007
- Manuscript Accepted: 2 MAY 2007
- Manuscript Revised: 30 MAR 2007
- Manuscript Received: 6 FEB 2007
- Fondazione Umberto DiMario
- Crohn's disease (CD);
- postoperative recurrence;
- ileocolonoscopy (CC);
- small intestine contrast ultrasonography (SICUS);
- wireless capsule endoscopy (WCE)
Background: The best available tool to assess recurrence of Crohn's disease (CD) is ileocolonoscopy (CC). Small intestine contrast ultrasonography (SICUS) and wireless capsule endoscopy (WCE) are noninvasive techniques able to detect small bowel lesions. In a prospective longitudinal study, we aimed to investigate the usefulness of SICUS and WCE for assessing postoperative recurrence of CD 1 year after surgery, using CC as the gold standard.
Methods: Twenty-two patients (11 men, median age 33 years, range 22–67 years) undergoing ileocolonic resection for CD were prospectively followed from July 2003 to May 2006, with the Crohn's Disease Activity Index (CDAI) used for clinical assessment every 3 months for 1 year. At 1 year, recurrence was assessed by SICUS and CC, followed by WCE. CD recurrence was assessed by CC (Rutgeerts score). SICUS was performed after ingestion of polyethylene glycol, and WCE was performed with Given M2A equipment.
Results: At 1 year, all 22 patients had inactive CD (CDAI < 150). In 5 patients, WCE was not performed because of luminal narrowing or stenosis. Seventeen of the 22 patients had all 3 techniques performed. CC detected recurrence in 21 of 22 patients. Lesions compatible with recurrence were detected by SICUS in all 22 patients (1 false positive). When considering only the 17 patients studied by all 3 techniques, recurrence was detected by CC in 16 of 17 patients, whereas lesions compatible with recurrence were detected by SICUS in all 17 patients (16 true positives [TPs], 1 FP) and by WCE in 16 of 17 patients (16 TPs, 1 true negative).
Conclusions: The present findings suggest that SICUS and WCE may be used as noninvasive techniques for the assessment of recurrence of CD in patients being regularly followed up after ileocolonic resection.
(Inflamm Bowel Dis 2007)
Postoperative recurrence after ileocecal resection is a feature of Crohn's disease (CD).1–4 Estimates of endoscopic recurrence in CD patients 1 year after resection vary from almost 73% to 95%, most often in the absence of overt symptoms (asymptomatic recurrence).1–6 The development of severe lesions in the early postoperative period, as observed by ileoscopy, has been associated with an earlier clinical relapse.2 Follow-up of CD patients after surgery includes clinical assessment. Ileocolonoscopy 1 year after resection, or even after 6 months, may help to evaluate recurrent symptoms and to define the type and severity of lesions.5 The gold standard for assessing CD recurrence currently includes conventional colonoscopy with ileoscopy.2 Noninvasive procedures are needed for assessing postoperative recurrence of CD. Several techniques have been investigated for this purpose (e.g., fecal calprotectin, fecal α1-antitripsin, scintigraphy using radiolabeled leukocytes, virtual colonoscopy, ultrasonography), although none has proved to be comparable to conventional colonoscopy.7–13
Transabdominal ultrasonography has been proposed for use in detecting small bowel lesions in patients with suspected or known CD, in whom it has a sensitivity of 67% to 84% and 81% to 95%, respectively.14–20 The use of an oral contrast significantly increases the sensitivity of ultrasonography for assessing small bowel lesions in patients with a suspected or known diagnosis of CD (>95%).21–23 It has been suggested that small intestine contrast ultrasonography (SICUS) may visualize not only established CD lesions, but also minor changes in the small bowel wall.21, 22 Wireless capsule endoscopy (WCE) is a new diagnostic tool for imaging the entire small bowel.24 The limited length of traditional colonoscopes allows visualization only of the distal ileum, leaving most of the small bowel unseen by the endoscope. WCE has been reported to accurately detect small bowel lesions and thus has been proposed as a new noninvasive technique for detecting CD lesions in the small bowel.25–27 However, major concerns about the use of WCE include the risk of capsule retention in patients with small bowel strictures.28, 29 Although studies have shown that both SICUS and WCE appropriately detect small bowel lesions, no studies have explored the usefulness of either technique in visualizing lesions related to postoperative recurrence of CD. Therefore, the primary aim of this prospective longitudinal study was to investigate the usefulness of SICUS and WCE for assessing recurrence in CD patients, using ileocolonoscopy as the gold standard. In particular, we aimed to assess whether WCE and SICUS can provide information on the presence and site (anastomotic or neoterminal ileum) of CD recurrence in patients undergoing ileocolonic resection that is comparable with that provided by ileocolonoscopy.
MATERIALS AND METHODS
From January 2004 to April 2005, 24 patients undergoing elective ileocolonic resection for CD (median age 34.5, range 22–67 years) were consecutively enrolled in a prospective longitudinal study. Of these 24 patients, 22 were prospectively followed up for 12 months, and 2 were lost to follow-up. Therefore, data from 22 of the 24 patients (91.6%) were analyzed. All patients were in regular follow-up at the GI Unit of the Università “Tor Vergata” of Rome, Italy, and were prescribed mesalamine (2.4 g/day) within 14 days of surgery. The diagnosis of CD was made according to standard clinical, endoscopic, histologic, and radiological criteria. Clinical characteristics of each patient are summarized in Table 1. In all patients, surgical resection included all the involved tissue, with no residual lesions at the resection margins. CD lesions before surgery involved the distal ileum in 19 patients (in 2 as a recurrence) and both the distal ileum and the right colon in 3 patients. At enrollment some patients had known risk factors for recurrence: 7 patients were smokers, and the indication for surgery of 5 patients was perforating subtype (Table 1). Clinical activity was assessed according to the Crohn's Disease Activity Index (CDAI) 3, 6, and 12 months after surgery.30
|Patient||Sex||Age||BMI||Surgical indication||Surgical access||Anastomosis||Number of resections|
|1. IM||M||67||28||Obstruction*||Laparatomy||Ileo-trasversus E-S||2|
|2. PM||F||45||27||Obstruction*||Laparotomy||Ileo-ascending S-S||1|
|3. RR||M||49||26||Obstruction*||Laparotomy||Ileo-ascending S-S||1|
|4. AL||M||34||23||Obstruction*||Laparatomy||Ileo-ascending E-E||1|
|5. GDG||M||28||29||E-C fistula||Laparatomy||Ileo-ascending S-S||1|
|6. MM||F||32||21||Abscess||Laparatomy||Ileo-ascending S-S||1|
|7. SR||F||39||23||Obstruction*||Laparotomy||Ileo-ascending S-S||1|
|8. AT||F||57||22||Obstruction*||Laparotomy||Ileo-trasversus S-E||1|
|9. AB||M||30||22||Perforation||Laparotomy||Ileo-ascending E-S||1|
|10. LF||F||32||20||Obstruction*||Laparoscopy||Ileo-ascending S-S||1|
|11. AD||F||54||22||Steroid dependent||Laparoscopy||Ileo-trasversus S-S||3|
|12. CP||F||35||20||Obstruction*||Laparoscopy||Ileo-ascending S-S||1|
|13. ADR||M||42||29||Obstruction*||Laparotomy||Ileo-ascending S-S||1|
|14. VGr||F||32||18||Steroid dependent||Laparoscopy||Ileo-ascending S-S||1|
|15. FF||F||29||20||Obstruction*||Laparoscopy||Ileo-ascending S-S||1|
|16. MC||F||47||18||Obstruction*||Laparoscopy||Ileo-ascending S-S||1|
|17. MDS||M||22||21||Anaemia||Laparotomy||Ileo-ascending S-S||1|
|18. VG||M||22||23||Obstruction*||Laparoscopy||Ileo-ascending E-S||1|
|19. SM||M||35||27||Obstruction*||Laparotomy||Ileo-ascending S-S||1|
|20. GZ||M||46||21||E-C fistula||Laparotomy||Ileo-ascending S-S||1|
|21. PV||M||44||21||Obstruction||Laparotomy||Ileo-ascending S-S||1|
|22. DM||M||27||23||Perforation||Laparotomy||Ileo-ascending E-S||1|
Inclusion criteria were CD patients under regular follow-up, age 18 to 70 years old, current elective ileocolonic resection, and written informed consent for the 3 procedures. Exclusion criteria (related to WCE) were previous intestinal resections for conditions other than CD, low compliance (poorly adherent to follow-up), diverticula, blind loop, pacemaker, or neurological disorders with altered gastrointestinal motility. The study was approved by the local ethics committee.
From January 2004 to April 2005 all eligible inactive (CDAI < 150) patients with no complications after ileocolonic resection were enrolled.
SICUS, WCE, and Colonoscopy at 12 Months
All 22 patients had a clinical assessment at 3, 6, and 12 months. At 12 months, all patients had SICUS, then ileocolonoscopy, and then WCE performed sequentially. WCE was not performed in patients showing strictures or stenosis by SICUS and/or endoscopy. Colonoscopy was chosen as the gold standard for CD recurrence, graded according to the Rutgeerts score (grades 0–4).2
SICUS and WCE at 3, 6, and 12 Months, Colonoscopy at 12 Months
Ten randomly selected patients had SICUS and WCE performed at 3, 6, and 12 months and colonoscopy at 12 months.
Subanalysis of Recurrent Lesions in Neoterminal Ileum
In the subanalysis, CD lesions limited to the anastomosis were not considered a recurrence.
SICUS was performed as described previously21, 23 after patients had ingested 375 mL (range 250–500 mL) of polyethylene glycol (PEG; Promefarm, Milano, Italy) as an oral contrast solution. SICUS was performed (Hitachi, EUB 6500, Japan) with 3.5- and 5-MHz convex and linear-array transducers. All the ultrasonographic procedures were performed by the same gastroenterologist (whose experience exceeds 2000 examinations), who was blinded to the WCE and colonoscopy results. A SICUS examination lasted a median of 40 minutes (range 35–90 minutes). Recurrence of CD was defined as finding23: (1) increased bowel wall thickness (>3 mm); (2) “stiff loop,” identified as a small bowel loop with increased bowel wall thickness not distended by contrast solution; (3) small bowel dilation, defined as a lumen diameter greater than 2.5 cm; (4) bowel stricture, defined as a lumen diameter less than 1 cm, measured at the level of maximally distended loop, independent of the presence of prestenotic dilation; (5) fistulae, defined as an hypoechoic tract with or without hyperechoic content; (6) mesenteric enlargement and/or masses; and (8) abscesses, identified as roundish anechoic lesions with an irregular wall, often showing internal echoes and posterior echo enhancement.
WCE examination was performed with a Given M2A capsule (Given Imaging, Yoqneam, Israel).24, 27 The evening before WCE examination, patients underwent bowel preparation with 2 L of PEG. Exclusion criteria for having WCE in addition to those already reported were strictures or stenoses determined by SICUS or colonoscopy. WCE images were assessed by the one gastroenterologist unaware of results from SICUS and colonoscopy. As there is not a standardized definition of CD recurrence as assessed by WCE, in accordance with previous studies,31 separate detection of ulcers, strictures, or stenosis in the neoterminal ileum and/or anastomosis was considered compatible with CD recurrence. The presence of erosions was also reported, although this alone did not determine a diagnosis of CD recurrence. Other lesions detected in the stomach and/or esophagus or in the upper small bowel (jejunum, proximal ileum) were defined as “upper” lesions but not considered CD recurrence. Any other lesion in the GI tract visualized by WCE was recorded and reported.
All colonoscopies were performed 1 year after surgery, with the degree of recurrence assessed according to Rutgeerts et al.2 All endoscopies were performed with the same colonoscope (diameter 11.3 mm) by the same gastroenterologist, who was blinded to the SICUS and WCE findings. Endoscopic findings in all patients were documented, and biopsy specimens were taken from the perianastomotic area for routine histological assessment.
The SICUS and WCE results are reported as being true positive (TP), true negative (TN), false positive (FP), or false negative (FN). Data are expressed as medians and ranges. Differences between groups were assessed by the Student t test. Correlation between endoscopic score and bowel wall thickness as assessed by SICUS was evaluated using the Yates coefficient of correlation.
SICUS, WCE, and Colonoscopy at 12 Months
No patients showed complications related to SICUS, WCE, or colonoscopy (Fig. 1). All 22 patients were in remission (CDAI < 150) at the 1-year follow-up. All patients were on mesalamine (5-ASA) within 1 month of surgery, whereas none used NSAIDs.
At 12 months, conventional ileocolonoscopy detected recurrence as defined by Rutgeerts score (1–4)2 in 21 of the 22 patients (95.4%). The endoscopy scores (shown in Table 2) were: grade 0, 1 patient; grade 1, 3 patients; grade 2, 7 patients; grade 3, 3 patients; and grade 4, 8 patients (with stenosis not passed by the endoscope in 1). No lesions apart from those indicating CD recurrence were detected in any of the 22 patients.
|Recurrence (Y/N)||Stenosis (Y/N)||Wall thickness (mm)||Recurrence (Y/N)||Stenosis (Y/N)||Upper Lesions (Y/N)||Recurrence (Y/N)||Stenosis (Y/N)||Grade (0–4)|
SICUS detected lesions indicative of recurrence in all 22 patients (1 FP, 21 TPs; sensitivity 100%, specificity 0%). The only patient with a FP finding showed only a mildly increased perianastomotic bowel wall thickness (3.5 mm), with no luminal narrowing and/or dilation. Median wall thickness in the perianastomotic area in patients with endoscopic recurrence was 5 mm (range 3.5–10 mm). Wall thickness in the perianastomotic area as assessed by SICUS (mm) was not highly correlated with endoscopic degree of recurrence in each of the 22 patients (r = 0.42, P = 0.05; Fig. 2). The difference in median bowel wall thickness between patients endoscopically showing grades 0 to 2 recurrence (n = 11) and patients endoscopically showing grades 3 or 4 recurrence (n = 11) was not statistically significant (median 4 mm, range 3.5–5 mm, versus median 5 mm, range 4–10 mm, respectively; P = 0.13). Perianastomotic luminal narrowing (<1 cm) was detected in 7 of the 22 patients; in 2 of these 7 patients it was associated with ileal loop dilation. Four of the 7 patients showed grade 4 recurrence, and 3 showed grade 2 recurrence. The 7 patients showing luminal narrowing also had a significantly thicker bowel wall in the perianastomotic area than did the 15 patients who did not show luminal narrowing (median 5, range 4–10 mm, versus median 4, range 3.5–5.0 mm, respectively; P = 0.029). As shown in Table 1, the body mass index (BMI), in particular, the obesity, of tested patients did not appear to influence SICUS findings (median 22, range 17–29).
At 12 months, only 17 of the 22 patients (77.2%) underwent WCE. Five patients did not, 4 because of luminal narrowing detected by colonoscopy (grade 4) and 1 because of a sonographic finding of both increased bowel wall thickness (5 mm) and stricture (Fig. 3). In the 4 patients with grade 4 recurrence, it was not easy to pass the endoscope through a perianastomotic luminal narrowing, suggesting that WCE not be done.
WCE findings compatible with recurrence were detected in 16 of 17 patients (94.1%; 1 TN, 16 TPs; 100% sensitivity and specificity). No patients showed stenosis or strictures. Additional WCE findings compatible with CD lesions in the small bowel above the distal ileum visualized by ileocolonoscopy were observed in 13 of 17 patients (76.4%), including the only patient with no recurrence. The upper GI lesions detected in the 13 CD patients included erosions and aphthoid ulcers surrounded by macroscopically normal mucosa, whereas no deep ulcers or strictures were detected. No other lesions (i.e., angiodysplasia, lymphomas, etc.) were detected in any of the 17 patients.
Considering only the 17 patients investigated by all 3 techniques at 12 months indicated that colonoscopy detected CD recurrence in 16 patients (94.1%), SICUS in 17 patients (16 TPs, 1 FP; 100% sensitivity; 0% specificity), and WCE in 16 patients (94.1%; 16 TPs, 1 TN; 100% sensitivity, 100% specificity). In the only patient who showed no recurrence, WCE showed no lesions (TN), whereas SICUS detected the bowel wall thickness as being mildly increased (3.5 mm), a false positive.
Subanalysis of Recurrent Lesions in Neoterminal Ileum
When those patients with recurrence confined to the anastomosis (with no involvement of the neoterminal ileum) were excluded, recurrence was detected in 19 of 22 patients (86%). The results showed that the degree of recurrence was grade 1 in 3 patients, grade 2 in 5 patients, grade 3 in 3 patients, and grade 4 in 8 patients (associated with stenosis in 1 patient). SICUS showed changes suggesting ileal recurrence in 19 of 22 patients (86%; 2 FNs, 2 FPs, 1 TN, and 17 TPs). The 2 false-negative findings were a patient with a grade 1 recurrence and a patient with a grade 3 recurrence. SICUS detected recurrence confined to the anastomosis and confirmed by endoscopy in both patients with no ileal recurrence (2 FNs) as detected by SICUS. Among the 17 patients undergoing WCE at 12 months, recurrence was detected in 14 of 17 (82%), including 1 FN, 1 FP, 2 TNs, and 13 TPs. No strictures or stenoses were detected in any of these patients. In 2 of the 17 patients, WCE did not visualize the anastomosis.
Considering only the 17 patients investigated by all 3 techniques showed that colonoscopy detected lesions in the neoterminal ileum in 14 patients (82%), SICUS in 14 patients (82%; 2 FNs, 2 FPs, 1 TN, and 12 TPs), and WCE in 14 patients (82%; 1 FN, 1 FP, 2 TNs, and 13 TPs). The only FN finding by WCE was a grade 2 endoscopic recurrence. Table 3 summarizes the sensitivity, specificity, positive predictive value, and negative predictive value for detecting CD recurrence by SICUS and WCE, when ileocolonoscopy (CC) was used as the gold standard. A comparison of WCE and SICUS showed that FN findings were observed in different patients, with 1 FN found by WCE (grade 2) and 2 FNs found by SICUS (grade 1 and grade 3). The false positives detected by WCE and SICUS also were found in different patients: a grade 2 recurrence limited to the anastomosis by the former and a grade 2 recurrence limited to the anastomosis and a grade 0 (no recurrence) by the latter. Similarly, TN findings by WCE and SICUS were observed in different patients: a grade 0 (no recurrence) and a grade 2 recurrence limited to the anastomosis by the former and a grade 2 recurrence limited to the anastomosis by the latter.
|Lesions compatible with recurrence||Sensitivity||Specificity||PPV||NPV|
|SICUS (limited to neoterminal ileum)||86%||33%||86%||33%|
|WCE (limited to neoterminal ileum)||93%||67%||93%||67%|
SICUS and WCE at 3, 6, and 12 Months, Colonoscopy at 12 Months
Ten patients underwent SICUS and WCE at 3, 6, and 12 months and ileocolonoscopy at 12 months (Table 4). At 3 months, changes suggesting recurrence at the anastomosis only or at both the anastomosis and the neoterminal ileum were detected by SICUS in 4 of the 10 patients. No patients showed luminal narrowing or dilation. Luminal lesions compatible with recurrence were detected by WCE in 9 of the 10 patients, in 4 of whom they were also visualized by SICUS. Upper lesions were detected in 1 patient, also showing lesions compatible with recurrence. At 6 months, 10 patients underwent SICUS, whereas 9 underwent WCE. SICUS and WCE detected lesions compatible with recurrence in 8 patients. Luminal narrowing was observed by SICUS in 1 patient, who was therefore excluded from WCE analysis. No patients showed bowel dilation. WCE showed upper lesions in 5 patients, all with recurrence. At 12 months, ileocolonoscopy showed recurrence in 9 of the 10 patients (grade 0 = 1, grade 1 = 0, grade 2 = 4, grade 3 = 1, grade 4 = 4; Fig. 4). Changes suggesting recurrence were detected by SICUS in all 10 patients (9 TPs, 1 FP). Luminal narrowing was detected in 3 patients, in 2 of whom it was associated with bowel dilation. Five patients were excluded from having WCE on the basis of the SICUS and/or colonoscopy findings of luminal narrowing or severe recurrence. Lesions compatible with recurrence were detected by WCE in 4 of these 5 patients (4 TPs, 1 TN). All 4 true positives found by SICUS were compatible with recurrence. No other lesions (angiodysplasia, lymphomas, etc.) were detected by WCE at 3, 6, and 12 months.
|Technique||3 Months||6 Months||12 Months|
|SICUS||4/10||8/10||10/10 (9 TPs,1 FP)|
|WCE||9/10||9/9*||4/5† (4 TPs, 1 TN)|
Subanalysis of Recurrent Lesions in the Neoterminal Ileum
At 3 months, SICUS showed lesions compatible with ileal recurrence in 3 of the 10 patients. No patients showed luminal narrowing or dilation. Recurrence was detected by WCE in 7 of the 10 patients and also by SICUS in 3 of them. Upper lesions were detected in 1 patient, also showing lesions compatible with recurrence. At 6 months, SICUS showed recurrence at the neoterminal ileum in 7 of the 10 patients. Luminal narrowing was detected in 1 patient, whereas none showed bowel dilation. WCE was performed in only 9 patients, with luminal narrowing the reason that 1 patient did not have WCE. Lesions compatible with recurrence were detected in 7 of the 9 patients. Upper lesions were detected in 5 patients, with 3 of them also showing lesions compatible with recurrence. At 12 months, ileocolonoscopy showed recurrence at the neoterminal ileum in 8 of the 10 patients (grade 0 = 2, grade 1 = 0, grade 2 = 3, grade 3 = 1, grade 4 = 4), whereas SICUS showed lesions compatible with recurrence in 9 of these 10 patients (8 TPs, 1 TN, 1 FP). Luminal narrowing was detected in 3 of these 9 patients, in 2 of whom it was associated with bowel dilation. Lesions compatible with recurrence were detected in 4 of the 5 patients who underwent WCE (3 TPs, 1 TN, 1 FP). All 3 true positives found by WCE also showed lesions compatible with CD recurrence by SICUS. No lesions (angiodysplasia, lymphomas, etc.) were detected by WCE at 3, 6, and 12 months.
In a prospective longitudinal study, we investigated the usefulness of SICUS and WCE for assessing the postoperative recurrence of CD when using ileocolonoscopy as the gold standard in a cohort of patients followed up for 1 year after surgery. An analysis of using SICUS and WCE for visualizing lesions compatible with CD recurrence in the early postoperative period (3 and 6 months) was also done. To our knowledge, this is the first prospective study to investigate the usefulness of these 2 noninvasive techniques for assessing CD recurrence. Recently, a prospective study compared ileocolonoscopy with WCE for detecting CD recurrence, limited to the early postoperative period (6 months).31 Results from a prospective study by Bourreille et al indicated that at 6 months the sensitivity of WCE in detecting CD recurrence in the neoterminal ileum was inferior to that of ileocolonoscopy, although WCE detected lesions outside the scope of ileocolonoscopy in more than two thirds of the patients.31 However, unlike in our study, Bourreille et al did not search by SICUS for lesions compatible with recurrence.31
In the present study, a high proportion of patients showed endoscopic recurrence (95%) at 12 months, which was within the expected range (73%–95%).1–4 All patients received mesalamine after surgery, maintaining clinical remission at 12 months. There have been conflicting findings on whether mesalamine is efficacious in preventing CD recurrence.32–38 Risk factors for early recurrence include having a perforating subtype and being a smoker.1–4, 39–41 In our population, 23% of patients showed a perforating subtype, and 32% were smokers, indicating that almost half the patients (10 of 22; 45%) had at least 1 of these 2 risk factors. The prevalent pattern of lesions in each patient was comparable before and after surgery, confirming our previous findings.42
Transabdominal ultrasonography (TUS) is a valuable noninvasive and radiation-free tool in the diagnosis of small bowel CD.14–20 However, in clinical practice TUS is not widely used for assessing small bowel lesions because of the virtual lumen and the presence of gas in the intestinal loops.21–23 To overcome the inherent limitations of TUS, SICUS was developed by using an oral contrast anechoic fluid. SICUS has been shown to be able to dissociate the small bowel loops, thus allowing measurement of bowel wall thickness and the lumen diameter at different levels.21, 23 In experienced hands, SICUS has indeed been shown to detect intestinal lesions in patients with suspected small bowel diseases with high (>95%) sensitivity and specificity, when compared with small bowel follow-through and small bowel enema.21 SICUS has been shown to be more accurate than TUS in assessing small bowel CD lesions, although the experience of the sonologist significantly affects the accuracy of both techniques, particularly of TUS.21
In the present study, lesions compatible with recurrence were appropriately detected by SICUS in all 21 patients with endoscopic recurrence at 12 months. This finding suggests that SICUS may be a useful noninvasive technique for detecting ileal CD recurrence. In 1 patient, SICUS showed findings compatible with recurrence, not confirmed by colonoscopy. In this patient, bowel wall thickness was only slightly increased (3.5 mm; normal value = 3 mm), suggesting that values close to this cutoff should be a alert for possible false-positive findings. Differently, the other 2 patients with a bowel wall thickness of 3.5 mm showed endoscopic recurrence (both grade 1). The cutoff of 3 mm was chosen based on the currently available data.21, 43 Ileocolonoscopy and SICUS provide a different view of the ileum, as endoscopy as WCE visualizes the inner surface, whereas SICUS visualizes the bowel wall layers. Although this may suggest that SICUS and endoscopy provide discrepant findings when assessing CD lesions, in our study SICUS detected increased bowel wall thickness in all 21 patients with recurrence. We also found that the correlation between bowel wall thickness and endoscopic degree of recurrence was at the limit of statistical significance, further suggesting that SICUS may be a useful noninvasive technique for assessing ileal CD recurrence.
Although changes in bowel wall thickness may be a result of normal early postoperative changes, several observations support that this potential confounder did not significantly influence the sensitivity and specificity of SICUS in assessing CD recurrence. Among the 10 patients longitudinally studied by SICUS at 3, 6, and 12 months, the number of patients with sonographic findings compatible with recurrence did indeed increase over the follow-up period (detected in only 4 patients at 3 months, in 8 patients at 6 months, and in all 10 patients at 12 months). This finding suggests that a nonspecific increased perianastomotic bowel wall thickness in the early postoperative period did not significantly influence our results. Supporting this view, a study that used transabdominal ultrasound (with no oral contrast) showed that bowel wall thickness did not increase in patients with previous ileocolonic resection for cancer of the right colon.13
WCE has been shown to allow direct evaluation of the small bowel mucosa in CD.25–27, 31 A recent meta-analysis comparing the yield of WCE to other diagnostic modalities in patients with suspected or established nonstricturing small bowel CD44 indicated that WCE is superior to all other modalities for diagnosing nonstricturing small bowel CD, with 3 the number needed to treat (NNT) to yield 1 additional diagnosis of CD over small bowel barium radiography and 7 the NNT over colonoscopy with ileoscopy.44
Sixteen true positives and 1 TN were observed when using WCE, suggesting this technique may be useful for assessing ileal CD recurrence. However, 5 patients did not have WCE performed because of luminal narrowing. This finding indicates that at 12 months the presence of CD strictures needs to be excluded using conventional techniques in order to avoid WCE impact. WCE did not visualize the anastomosis in 2 of the 17 patients, suggesting that this technique may not be useful for visualizing CD recurrence limited to the anastomosis. SICUS and WCE concordantly detected 16 TPs findings at 12 months, whereas in 1 patient a FP finding was detected by SICUS but not by WCE. These 2 techniques indeed offer different views of the small bowel, including the inner surface when using WCE and the extraluminal surface when using SICUS. SICUS, but not WCE, may also visualize the extent of lesions and fistulae or abscesses.21–23, 45–47
As WCE did not detect the anastomoses in 2 patients and an increase in anastomotic wall thickness may be a result not of recurrence but of the surgical procedure itself, we also analyzed results by only considering lesions involving the neoterminal ileum as recurrence (excluding lesions limited to the anastomosis). According to this analysis, endoscopic recurrence was detected in a smaller proportion of patients (86%), whereas FP and FN findings were detected by SICUS and WCE in different patients.
Our finding that at 12 months SICUS showed lesions compatible with recurrence in all patients may suggest that this technique overestimates recurrence. This hypothesis was not supported by our findings that at 3 months only 4 of the 10 patients showed sonographic lesions compatible with recurrence. In the same patients, WCE at 3 months detected lesions compatible with recurrence in 9 of the 10 patients. In the same study, at 6 months lesions compatible with recurrence were detected in a larger proportion of patients by both SICUS and WCE. Among the 10 tested patients, 8 showed sonographic evidence of recurrence. One of the 8 patients showed a marked increase in bowel wall thickness; therefore, WCE was not performed. This finding further supports the need for careful selection of patients eligible for WCE as early as 6 months after ileal resection for CD in order to avoid the impact of WCE. Among the 9 patients examined by WCE at both 3 and 6 months, only 1 showed upper lesions at 3 months, persisting at 6 months. Among the remaining 8 patients, 4 developed upper lesions at 6 months. The small number of patients tested by WCE at 12 months (n = 5) does not allow any statement about the course of upper lesions detectable by WCE during the first year after resection for CD. These findings also indicate that, differently than at 12 months, lesions compatible with recurrence in the early postoperative period (i.e., at 3 and 6 months) may be detected in a higher proportion of patients by WCE than by SICUS.
Results from this first prospective longitudinal study investigating the usefulness of SICUS and WCE in assessing the postoperative recurrence of CD suggest that these noninvasive techniques may provide a useful tool in clinical management of patients in regular follow-up after ileocolonic resection for CD. However, the presence of strictures also needs to be excluded by conventional radiology or ultrasonography in the early postoperative period in order to avoid WCE retention in patients with previous intestinal resection for CD.
The authors thank Graziano Bonelli for his technical support.
- 13Role of ultrasonography in the diagnosis of postsurgical recurrence of Crohn's disease. Am J Gastroenterol. 1998; 93: 1117–1121., , , et al.Direct Link:
- 19A prospective evaluation of high resolution sonography in the diagnosis of inflammatory bowel disease. Eur J Gastroenterol Hepatol. 1992; 4: 173–182., , .
- 33Placebo-controlled clinical trial of mesalazine in the prevention of early endoscopic recurrences after resection for Crohn's disease. Groupes d'Etudes Therapeutiques des Affections Inflammatories Digestives (GETAID). Eur J Gastroenterol Hepatol. 1996; 8: 229–233., , , et al.
- 44A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with non-stricturing small bowel Crohn's disease. Am J Gastroenterol. 2006; 101: 954–964., , , et al.Direct Link: