Eur Radiol (2008) 18: 2530–2534DOI 10.1007/s00330-008-1042-6 GASTROINTESTINAL

Tae-Hyung KimHo-Young SongJi Hoon ShinIn Kook ParkJin Hyoung KimJin-Oh LimKyung Rae KimEugene K. Choi

Received: 31 August 2007Revised: 12 February 2008Accepted: 29 March 2008Published online: 4 June 2008# European Society of Radiology 2008

Usefulness of multifunctional gastrointestinalcoil catheter for colorectal stent placement

Abstract The purpose of this studywas to evaluate the usefulness of amultifunctional gastrointestinal coilcatheter for stent placement in 98patients with colorectal strictures. Thecatheter was used in 98 consecutivepatients for stent placement in therectum (n=24), recto-sigmoid (n=13),sigmoid (n=38), descending (n=6),transverse (n=11), splenic flexure(n=3), hepatic flexure (n=2), andascending (n=1) colon. The catheterwas made of a stainless steel coil(1.3 mm in inner diameter), a 0.4-mmnitinol wire, a polyolefin tube, and ahemostasis valve. Usefulness of thecatheter was evaluated depending onwhether the catheter could pass astricture over a guide wire andwhether measurement of the stricturelength was possible. The passage ofthe catheter over a guide wire beyondthe stricture was technically successful

and well tolerated in 93 (94.9%) of 98patients. In the failed five patients, itwas not possible to negotiate the guidewire due to presence of nearly com-plete small bowel obstruction. Theaverage length of stricture was6.15 cm (range, 3 cm to 20 cm) inpatients with the colorectal stricture.There were no procedure-relatedcomplications. In conclusion, themultifunctional coil catheter seems tobe useful in colorectal stentplacement.

Keywords Colon . Interventionalprocedure . Gastrointestinal

Introduction

Colorectal stent placement is safe and effective for thepalliative nonsurgical therapy or preoperative decompres-sion of malignant colorectal obstruction [1–8]. Recently,investigators have improved this increasingly used tech-nique to offer a safer option for the treatment of malignantcolorectal obstruction [3–8]. In fluoroscopically colorectalstent placement, passage of a guide wire through a strictureis a major obstacle to the success of the procedure. Althoughmost interventional radiologists use angiographic cathetersfor passing a guide wire in colorectal stent placement [1–3,5], they have some limitations stemming from the tortuouscurved angulations of the intestinal tract and intestinal folds[9, 10]. In addition, it is not possible to measure stricture

length without removal of the guide wire, and it is difficult toopacify the area of the stricture before stent placement.

To overcome the limitations of the angiographiccatheters in the colorectal tract, we used a multifunctionalgastrointestinal coil (MGC) catheter. The purpose of thestudy was to evaluate the usefulness of the added flexibilityto the MGC catheter in overcoming the acute anatomicalangulations and its ability to measure the length of stricturein 98 patients with a stricture of the colorectal tract.

Materials and methods

Informed consent was obtained from each patient, and thisretrospective study was approved by our institutional

T.-H. Kim . J. H. Shin . J. H. Kim .J.-O. Lim . K. R. KimAsan Medical Center, Radiology andResearch Institute of Radiology,Seoul, South Korea

I. K. ParkDongguk University, Life Science,Seoul, South Korea

E. K. ChoiWeill Medical College of CornellUniversity,New York, NY, USA

H.-Y. Song (*)Asan Medical Center – Radiology,388–1, Pungnap-2dong,Songpa-gu, Seoul, 138–736,Republic of Koreae-mail: hysong@amc.seoul.kr

review board. Between September 2001 and November2006, the MGC catheter (S&G Biotech Inc., Sung-nam,Korea) was used for stent placement in 98 consecutivepatients with obstruction of the colorectal tract (Table 1).Ninety-eight patients underwent colorectal stent placementfor palliative nonsurgical therapy (n=63) and preoperativedecompression of colorectal obstruction (n=35). The siteof strictures was the rectum in 24 patients, the recto-sigmoid colon in 13 patients, the sigmoid colon in 38patients, the descending colon in 6 patients, the transversecolon in 11 patients, the splenic flexure in 3 patients, thehepatic flexure in 2 patients, and the ascending colon in 1patient. There were 60 male patients and 38 female patients,with a mean age of 58.6 years (age range, 17 – 88 years).

Description of the MGC catheter was detailed in aprevious study [11]. In brief, the MGC catheter consisted ofa distal uncovered and middle covered coil part, and aproximal homeostasis valve part (Fig. 1). Two differentlengths of MGC catheter were used: an 80-cm-lengthcatheter, which was used from the rectum to the rectosig-moid colon, and a 150-cm-length catheter, which was usedfrom the descending colon to the ascending colon. The coilcatheter was constructed from a medical grade stainlesssteel coil (1.3 mm in inner diameter and 1.8 mm in outerdiameter), a 0.4-mm nitinol wire, a polyolefin tube, and ahemostasis valve.

The distal uncovered part was designed to be soft andflexible in order to allow for easy advancement over theguide wire. The middle-covered coil part was made of astainless steel coil and covered with a polyolefin tube.Seven radiopaque markers made of gold plates (0.1 mm inthickness and 1.0 mm in width) were attached to the distalend of the middle-covered coil part and multiple smallround holes (from 0.8 mm to 2.2 mm) between theradiopaque markers for spraying contrast medium. The

proximal end of the middle covered coil part was connectedto the hemostasis valve. The hemostasis valve partconsisted of a main arm with a hemostasis valve and aside arm without a hemostasis valve. Contrast medium wasinjected through the side arm of the hemostasis valve. Toprovide a torque to the MGC catheter for the colorectaltract, a 0.4-mm nitinol wire (65-cm length) was attachedalong the outside of the stainless steel coil 65 cm from theproximal end of the middle covered coil part (Fig. 1).

For colorectal stent placement, an enema tube wasintroduced into the rectum. A small amount of nonioniccontrast medium was injected through the tube to outlinethe stricture. Under fluoroscopic guidance, a hydrophilicguide wire (Radiofocus M; Terumo, Tokyo, Japan) wasinserted through the anus to the narrowed bowel segment.The MGC catheter was then advanced over the hydrophilicguide wire as proximal to the narrowing point as possible.To measure the length of the stricture, a coil catheter waspassed over the hydrophilic guide wire, and a small amountof nonionic contrast medium was injected through the sidearm of the proximal hemostasis valve part. The exchangeguide wire was replaced with a super stiff guide wire(Boston Scientific, Watertown, MA), and theMGC catheterwas then removed. We performed stent placement throughthe super-stiff guide wire under fluoroscopic guidance.

Figure 2 describes the flow chart used in evaluating theusefulness of the MGC catheter. At first, we passed theMGC catheter through the orifice of the stricture underfluoroscopic guidance. In cases of technical failure innegotiation of the coil catheter or exchange guide wirethrough the stricture under fluoroscopic guidance, com-bined endoscopic and fluoroscopic guidance was attemptedwith or without a guiding sheath. The guiding sheath wasdesigned to allow for easy introduction of the MGCcatheter by overcoming the acute anatomical angulations ofthe gastrointestinal tract. The construction of the guidingsheath was described in detail by Bae et al. [12]. Theendoscope or guiding sheath was removed with the guide

Table 1 Summary of 98 patients with colorectal obstruction

Location No. of patients

Rectum 24

Recto-sigmoid colon 13

Sigmoid colon 38

Descending colon 6

Transverse colon 11

Splenic flexure 3

Hepatic flexure 2

Ascending colon 1

Indication No. of patients

Palliative Tx 63

Preoperative Tx 35

Tx = Therapy

Fig. 1 Schematic diagram of MGC catheter. The MGC catheterconsisted of a distal uncovered coil part, middle covered part, andproximal hemostasis valve part

Fig. 2 Evaluation of the usefulness of MGC catheter. MGC =multifunctional gastrointestinal coil; sheath = guiding sheath; endo =endoscopy

2531

wire left in place, and the MGC catheter was thenintroduced over the guide wire to the area of interestunder fluoroscopic guidance. After an exchange guide wirewas introduced through the working channel of theendoscope into the orifice of the obstruction, negotiationof the guide wire through the obstruction into the colon wasperformed under combined endoscopic and fluoroscopicguidance.

The technical success rate, the causes of technicalfailure, and complications were analyzed. The technicalsuccess was defined by whether it could pass the strictureover a guide wire and whether the measurement of thestricture length was possible by opacifying the proximaland distal parts of the stricture while the guide wire was inplace. In addition, we also measured the length of thestricture.

Results

There were no procedure-related complications. Thepassage of the MGC catheter over an exchange guidewire beyond the stricture under fluoroscopic guidance wastechnically successful and well-tolerated in 93 patients

(95%). Stent placement was successful in all 93 patientswho underwent successful negotiation of the coil catheterthrough the stricture. In the remaining five patients, wewere not able to pass the guide wire through the obstructionsite despite combined fluoroscopic (n=3) and combineduse of a guiding sheath (n=2).

Sixty-eight (73%) of 93 patients underwent successfulpassage of the guide wire under fluoroscopic guidance(Fig. 3), 16 patients (17%) under guiding sheath andfluoroscopic guidance (Fig. 4), 6 patients (7%) undercombined endoscopic and fluoroscopic guidance (Fig. 5),and 3 patients (3%) under combined endoscopic, guidingsheath, and fluoroscopic guidance. In the five failedpatients (recto-sigmoid: 1, sigmoid colon: 2, descendingcolon: 1, transverse colon: 1), it was not possible tonegotiate the guide wire through the obstruction due to thenear-complete obstruction.

The total average length of stricture was 6.15 cm (range,3 cm to 20 cm) in patients with successful opacification ofthe proximal and distal parts of the stricture (Table 2).Average stricture lengths were 6.39 cm (range, 3 cm to12 cm) in patients with stricture at the rectum, 5.27 cm(range, 3 cm to 9 cm) in patients with stricture at the recto-sigmoid colon, 5.84 cm (range, 3 cm to 20 cm) in patients

Fig. 3 Radiographs obtainedduring stent placement in apatient with advanced gastriccancer involving the transversecolon. a Injection of contrastmedium through a MGC cathe-ter while the guide wire (curvedarrows) is in place. The strictureappears as a filling defect (ar-rows). b Barium study obtained3 days after stent placementshows good flow of contrastmedium through the stent(arrows)

Fig. 4 Radiographs obtainedduring stent placement in apatient with rectal cancer. a Theplain radiogram shows difficultpassage of the wire through thestricture (arrow) at the rectosig-moid colon. b The guide wire(curved arrow) was passedunder combined guiding sheath(arrow heads) and fluoroscopicguidance. Injection of contrastmedium through a MGC cathe-ter (arrows) shows a stricture

2532

with stricture at the sigmoid colon, 5.8 cm (range, 4 cm to10 cm) in patients with stricture at the descending colon,6.63 cm (range, 3 cm to 14 cm) in patients with strictureat the transverse colon, 5.0 cm (range, 3 cm to 8 cm) inpatients with stricture at the splenic flexure, 5.5 cm inpatients with stricture at the hepatic flexure, and 4.0 cmin patients with stricture at the ascending colon.

Discussion

Expandable metallic stents are being increasingly used inpalliative nonsurgical therapy or preoperative decompres-sion for malignant colorectal obstructions [1–8]. Before astent delivery system is passed through a guide wire, closeevaluation of stricture length and severity is essential, andsuccessful advancement of the guide wire is mandatory.Despite the traditional use of angiographic catheter innegotiating the guide wire through the obstruction site, it isdisadvantaged by their limitations in colorectal stent place-ment [4, 5]. As well as its inability to be pushed to overcome

acute angulations of the colorectum and inability to carry outone-step measurement of the stricture length withoutremoving the wire [13], the angiographic catheters havedifficulty controlling a guide wire due to friction between theinner lumen of the catheter and the outer surface of the guidewire [11, 14]. Furthermore, its short length (maximum110 cm) and inconducive shape make it difficult reach distallesions from the rectum. Therefore, the procedure time maybe longer and the chance of technical failure for colorectalstent placement may be even greater. For instance, weachieved 95% (93/98) technical success including 17% ofdistal lesions (transverse colon to ascending colon). On thecontrary, a report using the angiographic catheter forfluoroscopic colorectal stent placement achieved 89%(33/37) of technical success with no distal lesions [13].

Developed specifically for non-vascular interventions,MGC catheter has some advantages in colorectal stentplacement. First, it allows for easy injection of contrastmedium. With a MGC catheter, contrast medium can beinjected with the guide wire in place in the catheter lumenbecause of the larger inter-space between the catheter andguide wire. The inter space between inner luminal diameterof the sizing catheter(Cook, Bloomington, IN) and a guidewire is 0.07 mm; a MGC catheter is 0.4 mm. Second, itscharacteristics make it easy tp adapt in the colorectal tract.In the general angiographic catheter, the total constructionis rigid and has less flexibility because it is made by rigidthermosetting resin and stainless steel mesh. Thesecharacteristics often make it difficult to pass vascularcatheters and guide wires into the colorectal tract becausetheir distal tips tend to become wedged in the tortuouscurved angulations of the intestinal tract and intestinalfolds. These problems of the angiographic catheter can beovercome with use of the MGC catheter, which enhancestorsion control in curved angulations of the intestinal tractand provides maximum flexibility. The distal tip of theMGC catheter is uncovered so it can provide flexibility to

Fig. 5 Radiographs obtained during stent placement in a patientwith rectal cancer involving the rectosigmoid colon. a Water-solublecontrast medium study before stent placement shows a completestricture (arrow) at the rectosigmoid colon. b The guide wire (curved

arrow) was subsequently passed under combined endoscopic andfluoroscopic guidance. c Injection of contrast medium through aMGC catheter while the guide wire (curved arrows) is in place. Noteopacification of the stricture (arrows)

Table 2 Stricture length of 98 patients with colorectal obstruction

Location Average stricture (range) length (cm)

Rectum 6.39 (3 – 12)

Recto-sigmoid colon 5.27 (3 – 9)

Sigmoid colon 5.84 (3 – 20)

Descending colon 5.8 (4 - 10)

Transverse colon 6.63 (3 – 14)

Splenic flexure 5.0 (3- 8)

Hepatic flexure 5.5

Ascending colon 4

Total 6.13 (3 – 20)

2533

the mucosa, which can minimize the risk of mucosal injury.The nitinol wire in the proximal 65 cm of the coil catheterimproves the handling characteristics of this catheter.

Although there were no major complications orprocedural problems, there is room for further improve-ment. We envision the next generation of the MGC catheteras having varied shapes that conform to different types ofobstruction. In patients with near complete or completeobstruction of the colorectal tract, for example, weexperienced some difficulties introducing the tip of the

MGC catheter through the stricture largely due to the tip’sflatness. We believe we can address this limitation byadopting the tip of a Headhunter or Cobra catheter. Inconclusion, the MGC catheter is useful in stent placementin patients with colorectal strictures.

Acknowledgments The study was supported by a grant (#HMP-00-B-31400-00169) from the Highly Advanced National Project,Ministry of Health and Welfare, Republic of Korea.

References

1. De Gregorie MA, Mainar A, Tejero E etal (1998) Acute colorectal obstruction:stent placement for palliative treatment-results of a multicenter study. Radiolo-gy 209:117–120

2. Mishima K, Sawada S, Tanigawa N etal (1999) Expandable metallic stenttreatment for malignant colorectalstrictures. Cardiovasc Interv Radiol22:155–158

3. Miyayama S, Matsui O, Kifune K et al(2000) Malignant colonic obstructiondue to extrinsic tumor: palliative treat-ment with a self-expanding nitinolstent. AJR Am J Roentgenol 175:1631–1637

4. Baron TH, Rey J, Spinell P (2002)Expandable metal stent placement formalignant colorectal obstruction.Endoscopy 34(10):823–830

5. Kang SG, Jung GS, Cho SG et al(2002) The efficacy of metallic stentplacement in the treatment of colorectalobstruction. Korean J Radiol 3:79–86

6. Keymling M (2003) Colorectal stent-ing. Endoscopy 35(3):234–238

7. Meisner S, Hensler M, Knop FK et al(2004) Self-expanding metal stents forcolonic obstruction: experiences from104 procedures in a single center. DisColon Rectum 47:444–450

8. Ptok H, Meyer F, Marusch F et al(2006) Palliatie stent implantation inthe treatment of malignant colorectalobstruction. Surg Endosc 20:909–914

9. Nether NF (1989) Atlas of humananatomy, The ciba collection of medicalillustrations. Novartis, East Hanover,New Jersey, pp 267–269

10. Woodburne RT, Burkel WE (1988)Essential of human anatomy, 8th edn.Oxford University Press, Inc., p. 472–480

11. Song HY, Shin JH, Lim JO et al (2004)Use of a newly designed multifunc-tional coil catheter for stent placementin the upper gastrointestinal tract. JVasc Interv Radiol 15:369–373

12. Bae JI, Shin JH, Song HY et al (2005)Use of guiding sheath in peroralfluoroscopic gastrointestinal stentplacement. Eur Radiol 15:2354–2358

13. Choi JS, Choo SW, Park KB et al(2007) Interventional management ofmalignant colorectal obstruction: use ofcovered and uncovered stents. Korean JRadiol 8:57–63

14. Shin JH, He X, Lee JH et al (2003)Newly designed multifunctional coilcatheter for gastrointestinal interven-tion: feasibility determined by experi-mental study in dogs. Invest Radiol38:796–801

2534