CT-guided percutaneous pelvic abscess drainage in Crohn’s disease

Background Percutaneous abscess drainage (PAD) is the current therapy for abdominal or pelvic collections. PAD has poorer curative rate for abscesses in Crohn’s disease (CD), commonly complicated by wide fistulas and multiloculations. Methods We

Please download to get full document.

View again

of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 0 | Pages: 7

Extension: PDF | Download: 0

  R. Golfieri (  ) • A. Cappelli • E. GiampalmaDepartment of RadiologyPoliclinico S. Orsola-MalpighiUniversity of BolognaBologna, ItalyE-mail: golfieri@aosp.bo.itF. Rizzello • P. Gionchetti • M. CampieriDepartment of Internal MedicinePoliclinico S. Orsola-MalpighiUniversity of BolognaBologna, ItalyS. Laureti • G. PoggioliDepartment of SurgeryPoliclinico S. Orsola-MalpighiUniversity of BolognaBologna, Italy tive success rates. In unsuccessful cases, due to wide fistu-las or postoperative anastomotic dehiscence, PAD helped inplanning elective surgery, reducing surgical complications. Key words Abscess, percutaneous drainage • Pelvicorgans, abscess • Pelvic organs, interventional procedures • Crohn’s disease Introduction Currently, percutaneous abscess drainage (PAD) is consid-ered the standard therapy for abdominal or pelvic infectedfluid collections due to its significant reduction of morbidityand mortality compared with surgical repair [1–11]. Agener- al consensus has been reached on the most important factorsconditioning the effectiveness of PAD: these are abscesscharacteristics (aetiology, unilocular or multilocular collec-tion, location and contact with vital structures, size and num-ber of collections, presence of fistulas), the patient’s clinicalconditions (APACHE score) and technical approach. Fistulasand multiloculations are common in Crohn’s disease (CD)abscesses and are usually considered predictors of poor out-come of PAD as a curative treatment [1, 3, 5–9, 12–14]. We undertook this retrospective study to evaluate thecurative results, complications and final outcomes of CT-guided PAD for pelvic collections in a selected populationof Crohn’s disease patients, considered at high risk fordrainage failure after PAD. Patients and methods Between January 1998 and January 2005, a total of 87 patients withCrohn’s disease (mean age, 36 years; range, 14–45 years) weretreated with PAD for pelvic abscesses. Seventeen patients had apostoperative collection 4–10 days after colorectal surgery (mean, 8 R. Golfieri • A. Cappelli • E. Giampalma • F. Rizzello • P. Gionchetti • S. Laureti • G. PoggioliM. Campieri CT-guided percutaneous pelvic abscess drainage in Crohn’s disease Received: 12 June 2005 /Accepted: 24 January 2006/ Published online: 19 June 2006ORIGINALARTICLE Abstract Background Percutaneous abscess drainage(PAD) is the current therapy for abdominal or pelvic col-lections. PAD has poorer curative rate for abscesses inCrohn’s disease (CD), commonly complicated by wide fis-tulas and multiloculations. Methods We retrospectivelyevaluated abscess cure rate, complications and final out-come in 87 CD patients, 70 with spontaneous and 17 withpostoperative pelvic abscesses, all treated with CT-guidedPAD during the last 7 years. Results A77% primary suc-cess rate and an 84.3% secondary success rate wereobtained without major complications. The higher successrate for PAD was for postoperative (88.2%) rather thanspontaneous abscesses (74.2%). Seventy-two percent of treated patients did not develop recurrent abscesses andunderwent elective surgery up to 40 months later. Conclusions PAD in pelvic abscess complicating CD is aneffective alternative to early surgery with satisfactory cura- Tech Coloproctol (2006) 10:99–105DOI 10.1007/s10151-006-0260-7  days), whereas 70 patients presented spontaneous abscess compli-cating active Crohn’s disease involving the colon-rectum or ileum. The abscesses were located in the right lower quadrant (17patients), within the iliacus, piriformis or iliopsoas muscles (18patients), in the right gluteus muscles (8 patients), or in the pre-sacral space (44 patients). The mean diameter of the abscesseswas 9 cm, ranging from 5 to 12 cm.PAD was performed under CTguidance. The transglutealapproach was most commonly used (71 of 87 patients: 81%),with the patient in prone decubitus position. The collection wasdrained via an anterior or lateral approach in only 16 patients,whenever the abscess was not too close to major pelvic vessels. Three different drainage techniques were used, according todepth, location and morphology of the collections. Direct trocartechnique was the most common (45 cases: 51.7%), inserting theassembled catheter directly. The tandem technique was adoptedin 23 cases (26.4%): a Chiba needle was inserted and used as aguide for the assembled catheter. The Seldinger technique wasused in 19 cases (21.9%) with multiloculated deep pelvic collec-tions: prior insertion of a sheath-needle set (Accustick model)was followed by definitive drainage (Fig. 1). Abscesses weredrained by an 8 French (F) single lumen catheter (APDLdrainage, Boston Scientific Medi-Tech) in 42 cases, 10 F APDcatheter in 23 cases and a 12 F catheter in 22 cases.After placement of the catheter, systemic antibiotic therapywas given until signs and symptoms of infection abated (reduced100fever and leukocytosis) and catheters were kept patent by irrigationwith 10–20 ml saline solution two to four times a day, dependingon the viscosity of fluid drained. Follow-up consisted of weeklyCTscans to determine the size of the cavity and to disclose sinustracts. An additional fluoroscopic control was performed after CTscan in patients with PAD complicated by persistent drainage oronly partial defervescence, injecting contrast media within the col-lection through the drainage (fistulogram) to better depict a pre-existing wide fistula associated with the bowel (Fig. 2). Catheterswere removed when drainage was absent for at least 3 consecutivedays, provided that CTconfirmed complete abscess resolution.For the purpose of reporting the results, the primary successof PAD was defined as clinical improvement (defervescence,decreased white blood cell count) and complete abscess drainagedocumented by imaging, followed by catheter removal withoutneed for surgery within 30 days of catheter removal. Secondarydrainage success was considered when the same results wereobtained after catheter repositioning or when multiple catheterswere required to drain multiloculated abscesses. Results All 87 abscesses were successfully drained at the firstattempt under CTguidance with no failures of primary R. Golfieri et al.: CT-guided abscess drainage in Crohn’s disease Fig. 1a-e PAD: Transgluteal approach, Seldinger technique . a Axial CTimage showing two postoperative fluid collections within the pre-sacral space: a prior 21-gauge Chiba needle aspiration confirmed the infected nature of the median collection and contrast medium injec-tion demonstrated absence of communications between the two collections. Aspiration of the second collection demonstrated a serousovarian cyst content ( b ). The abscess was then drained by a sheath needle (Seldinger) method: a 0.018-inch guidewire was inserted intothe needle ( c ); the needle was removed and replaced by a plastic sheath through which a 0.038-inch guidewire was introduced ( d ). Thefinal drainage in place, as documented in the CTscout view control ( e ) adebc  drainage positioning, major complications or enteric-cuta-neous fistulas within the course of PAD. Of the 87drainages, 67 (77%) were classified as primary successes,including 52 on spontaneous abscesses and 15 in postop-erative collections (Table 1). Defervescence was alwaysobtained in 24–48 h and catheter output usually decreasedin 2–7 days. The average duration of drainage was 10.5days (range, 3–23) with an average hospital stay of 14.6days (range, 11–48). Elective bowel resection of the dis-eased segments was thus allowed in these patients, with anoperation-free interval of at least 3 months.Twenty patients (23%) were classified as failures of pri-mary drainage. Among these, 7 spontaneous collectionsrequired catheter repositioning or multiple drainages.Persistent fever, suggesting undrained pus, led to promptrepeat CTscans revealing catheter dislodgment in 3 cases(it was repositioned under CTguidance during the samesession) and satellite collections in 4 cases (1–2 additionalcatheters were then inserted at the end of the same CTses-sion). Complete clinical resolution after prolongeddrainage was obtained in all these 7 patients. The averageduration of drainage was 23.8 days (range, 15–25), with amean hospital stay of 25 days (range, 13–40). These 7 cases(8%) needed a new or repeated drainage and were thereforeconsidered primary procedure failures but secondary pro-cedure successes, since early surgery was avoided in allcases. Considering both primary and secondary successes,an overall success rate of 85% was obtained.PAD definitively failed as a curative treatment, andsurgery related to the abscess or its underlying cause wasrequired in 13 cases (15%). Afollow-up CTscan and a fis-tulogram under fluoroscopic control in 11 spontaneousabscesses demonstrated a wide associated fistula to thebowel, indicated for temporized surgery which was per-formed within 30 days. Awide dehiscence of the surgicalanastomosis was present indicating immediate surgery in 2post-operative PAD cases with continuous drainage of intestinal content. R. Golfieri et al.: CT-guided abscess drainage in Crohn’s disease101 Fig. 2a-e Failures of primary abscess drainage . a Axial CTimage shows a multiloculated right pelvic abscess caused by multiple inter-nal fistulas. After catheter positioning ( b , c ), a fistulogram under fluoroscopy ( d ) demonstrated wide communications with the bowel loops,ascending and sigmoid colon, also confirmed at CTstudy ( e ) adebcTable 1 Results of percutaneous abscess drainage in 87 patients with Crohn’s diseasePrimary Secondary Failure, n (%)Global Recurrence, n (%)success, n (%)success, n (%)success, n (%)Postoperative collections (n=17)15 (88)0 1(0)2 1.(12)15 (88)0 11.(0) Spontaneous abscesses (n=70)52 (74)7 (100)11 (15.7)59 (84)11 (15.7)Total67 (77)7 1(8)13 1.(15)74 (85)11 (12.6)  Among the 17 postoperative collections, the successrate was higher (88%) since PAD was completely curativein all but 2 cases (Fig. 2). On the contrary, PAD was pri-marily successful only in 52 of the 70 spontaneous pelvicabscesses (primary success rate, 74%) with a secondarysuccess rate of 84% (59 of 70) obtained after catheter repo-sitioning in 3 patients and additional catheter insertions in4 cases.Amean follow-up of 39.3 months (range, 12–84) wasobtained for all patients. During the follow-up, among the74 patients successfully drained, 63 (85.1%) did not devel-op recurrent abscesses: 17 of them are still awaiting anelective surgical procedure with a follow-up ranging from12 to 40 months, whereas the 46 remaining patients under-went surgical treatment within 18 months after successfulPAD. Therefore, the benefit of PAD has been maintainedin the long-term follow-up in 63 of 87 patients (72%).Eleven patients (12.6%) developed recurrence abscess-es, all arising within 6 months in spontaneous abscesses(15.7%) in abdominal locations distant from the primaryabscess. Five recurrences required immediate surgery dueto poor clinical conditions whereas 6 were submitted tosuccessful percutaneous re-drainage which allowed thesurgical procedure to be scheduled within 18 months. Discussion Unanimously accepted criteria for patient selection forsuccessful PAD are still lacking. General criteria are basedon the most suitable size and shape of the fluid collectionand the “surgical attitude” of different centres. Gerzof etal.’s historical strict selection criteria confined to “simpleabscess” reported a PAD success rate of 85%–93% [1].Extending the indications to “complex abscesses” (multi-loculated, associated with fistulae, splenic abscesses andinfected fluid collections), the same group demonstratedlower curative rates (up to 45%) [15], in contrast to othermore promising success rates (70%–88%) in complex col-lections [3, 16–18]. Currently, the American College of Radiology (ACR) Appropriateness Criteria for PAD of abscesses include both simple and complex abscesses, pro-vided that safe drainage routes, without involvement of organs or structures and without direct contact with majorblood vessels, are present [14]. PAD is indicated in casesof infected fluid collections larger than 3 cm, since insmaller collections simple diagnostic aspiration to obtaincultures of fluid sample is usually sufficient to evacuatethe fluid content and render PAD unnecessary (the “3 cm”rule) (Fig. 1) [19].Wide differences have been reported in PAD successrates among the general population, ranging from 70% to93%, and in complication rates (from 1% to 15%) andmortality rates (from 1% to 6%) due to the heterogeneous 102 characteristics of patients included in different series [1–6,14, 20]. Prospective randomized studies comparing PADand open surgical drainage (OSD) are also lacking: theonly comparative data available are from the two majorretrospective case-controlled studies of Olak et al. [10]and Hemming et al. [11], demonstrating similar results interms of successful treatment rates (PAD 70% and 93% vs.OSD 85% and 96%, respectively), morbidity (PAD 29%and 29% vs. OSD 40% and 26%, respectively) and mor-tality (PAD 11% and 12% vs. OSD 7.4% and 14%, respec-tively). The two treatments resulted in similar durations of hospital stay, and the global condition of patients(APACHE II scores) was the only prognostic factor of morbidity and mortality equally in PAD and OSD. Predictorsof successful PAD (success rate about 90%)are absent fistulas, first collection, small size and locationin the right lower quadrant, lesions situated in the periph-ery of the abdomen, uncomplicated access routes throughthe abdominal wall, uniloculated homogenous fluid col-lections, and postoperative abscess without primary intra-abdominal disease; in contrast, abscesses caused by inter-nal fistulas, multiloculated or multiple, with infected clotsimpossible to drain are strong predictors of unsuccessfuloutcomes [7–9, 20]. Unfortunately, the characteristics related to lower success rates are the hallmarks of abscesses in Crohn’s disease. They are frequently multi-ple, multiloculated and supplied by one or more fistulatracts in up to 75% of cases, since the mechanism of abscess formation is mainly due to transmural bowel wallinflammation and underlying luminal stenosis, resultingin perforation and extralumenal abscess formation. On theother hand, PAD in Crohn’s disease is now accepted as afundamental pre-operative temporizing manoeuvre toimprove the patient’s condition before a more limited or asingle-stage surgical procedure [21]. In fact, surgicalmanagement of Crohn’s disease abscesses is problematic:incision and drainages are definitive procedures in only21%–27% of patients because of the formation of entero-cutaneous fistulas. By contrast, enterocutaneous or ente-rocolonic fistulas are seldom secondary to PAD [22]. Ourexperience confirms no evidence of such complicationsarising within the course of PAD. Surgical en bloc resection of the abscess followed byimmediate anastomosis should be avoided because of thepossibility of abscess rupture during mobilization.Immediate re-anastomosis is not always feasible and iscontraindicated if gross contamination is encountered inthe operative field. Currently the majority of surgeonsprefers a two-stage procedure with drainage of the abscessfollowed by elective surgery to treat the chronically dis-eased bowel [22]. In addition, use of PAD will allowaggressive medical therapy with prolonged parenteralhyperalimentation and bowel rest, which have been advo-cated to improve the patient’s nutritional status and facili-tate decreased output from fistula tracks. R. Golfieri et al.: CT-guided abscess drainage in Crohn’s disease  Technical aspects such as selection of the most appro-priate access route and guidance methods are fundamentalsteps of the procedure and essential to prevent complica-tions. Many percutaneous approaches have been describedwith ultrasound- or CT-guided methods [23–28]. Anteriorand lateral approaches are technically the simplest, even if not always feasible because of interposed bowel and otherpelvic viscera. In the present report, only 16 cases weredrained via anterior and lateral approaches, whenever theabscess was not too close to major pelvic vessels, where-as the transgluteal approach was the most commonly usedfor deep pelvic abscesses (71 patients), positioning thepatient in prone decubitus, as initially reported by Butch etal. [27]. Spiral CThas to be considered the most effectiveguidance method for PAD of deep-seated collections withoverlying vital structures. Fluoroscopy during fistulogramis also useful at follow-up to assess the reduction in cavi-ty size from healing and the presence of internal commu-nications, and to permit catheter exchanges and reposi-tioning. We used fluoroscopic control during follow-upafter CTonly in cases of persistent drainage or only par-tial defervescence which confirmed and depicted betterthan CTa pre-existing wide associated fistula to the bowelin 11 cases (Fig. 2).Catheter-related problems such as drainage migration,inadequate drainage, catheter obstruction, and the need formultiple drainages are not uncommon [1, 2, 5, 6, 20]: a sudden decrease or cessation of fluid drainage may revealcatheter obstruction or kinking and, conversely, a suddenincrease or change in fluid (from pus to bowel content)should alert to fistulous communications with the bowel.In our cohort, a repeated CTscan in 7 patients with per-sistent fever (suggesting inadequate drainage) demonstrat-ed catheter dislodgement in 3 cases. The catheter wasrepositioned under CTguidance during the same diagnos-tic CTsession whereas satellite collections of undrainedpus in 4 cases were treated directly by inserting one to twoadditional catheters. Acomplete resolution was obtained inall these patients after prolonged (3–4 weeks) multi-drainage positioning and they were considered secondaryPAD successes (Fig. 3).In the treatment of abscesses of aetiologies differentfrom Crohn’s disease, complete success rates of about85%–90% have been reported [1–16] with partial success in only 7%–18%, mainly related to the presence of fistu-las. The reported PAD success rates of 28%–50% inspontaneous Crohn’s abscesses are much lower thanthose reported in non-Crohn’s disease [12, 19, 22]. Weobtained a primary success rate of 77% and, includingthe secondary successes, an overall success rate of 85%,with an operation-free interval of at least 3 months. Thesurprisingly higher primary success rates compared tothe literature may be due to both the operators’technicalexperience and to constant monitoring of drainage outputin close cooperation with surgical staff. PAD solved theacute problem of the abscess and a postponed single-stage operation was reserved for resection of the dis-eased bowel. R. Golfieri et al.: CT-guided abscess drainage in Crohn’s disease103 Fig. 3a-d Secondary PAD success fol-lowing multiple concomitant drainages .Upper extension of a previously treatedlower pelvic abscess involving ilio-psoas and gluteus muscles, withoutclear communication with the bowelloops ( a , b ). Two drainages were insert-ed during the same diagnostic CTses-sion ( c , d ), with complete clinical reso-lution after prolonged drainage abcd
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks