| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Interventional Radiology, Institut Gustave Roussy, 39 rue
Camille Desmoulins, 94805 Villejuif, France.
2 Department of Anesthesiology, Institut Gustave Roussy, 94805 Villejuif,
France.
Received January 28, 2002;
accepted after revision July 16, 2002.
Address correspondence to T. de Baere.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred fifty-six patients with cancer, who ranged in age from 3 months to 75 years (mean ± SD, 47 ± 18 years), were referred 290 ± 200 days (mean ± SD; range, 0-1202 days) after central venous device placement for retrieval of a fractured and embolized central venous device catheter (n = 100), retrieval of a guidewire embolized during placement of a central venous device (n = 2), repositioning of the migrated tip of a central venous device catheter (n = 38), and fibrin-sheath stripping (n = 16). All procedures were performed with the patient under local anesthesia on an outpatient basis, except for the eight pediatric patients.
RESULTS. Ninety-five of the 100 embolized catheters and both of the guidewires were successfully retrieved. Retrieval was preceded by repositioning the embolized catheter with a pigtail catheter in 48 of these cases. Most of the procedures were performed with standard vascular tools (loop snares and pigtail catheters); the use of more sophisticated devices (grasping forceps, baskets, or balloons) rarely overcame the failure of a loop snare. Repositioning a migrated catheter tip was achieved with a pigtail catheter in 32 of 38 attempts. Of the repositioned catheters, only 24 could be used. Most of the 11 repositioning and retrieval failures were encountered because the catheter lacked a free end. Fibrin-sheath stripping was always technically successful: all these catheters were patent at 3-month follow-up. No procedure-related complications occurred.
CONCLUSION. The endovascular approach is highly feasible, safe, and effective for the management of mechanical complications of central venous devices. It is probably advisable to reserve endovascular repositioning for port catheters that are cumbersome to exchange and to replace simple catheters.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nonsurgical techniques dedicated to the management of these central venous device mechanical complications have been widely described [5,6,7,8,9,10], but to our knowledge, no prior study has evaluated the feasibility, safety and efficacy of the endovascular approach in a large homogeneous cohort. We present our experience in the management of these mechanical complications over 8 years in a large cancer treatment center.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Embolized central venous device catheters were documented on a chest radiograph 0-1202 days (median, 199 days; mean ± SD, 290 ± 200 days) after placement of the device. Among the 100 patients with embolized central venous devices, 61 had undergone catheter placement at our institution, and 39 were referred from other institutions. The distal tip of the embolized catheter was located in the pulmonary artery (n = 26), right ventricle (n = 24), right atrium (n = 23), inferior vena cava (n = 7), superior vena cava (n = 17), hepatic veins (n = 2), or a thymic vein (n = 1). Of these 100 embolized catheters, the reason for embolization was clearly identified in 84, including section of the catheter in the costoclavicular space (n = 52), separation from the port (n = 22), an accidental section during retrieval (n = 7), loss during implantation (n = 3).
The location of the tip of the migrated catheter was documented on a chest radiograph in 38 patients 1-156 days (median, 62 days) after port-catheter placement. The catheter tip was located in the jugular vein (n = 19), subclavian vein (n = 5), azygos vein (n = 4), homolateral internal mammary vein (n = 3), contralateral internal mammary vein (n = 3), thyroid vein (n = 3), or vertebral vein (n = 1). The location of each catheter was confirmed on phlebography using the central venous device before the interventional procedure.
Fibrin-sheath obstruction was suspected as a result of low flow during perfusion or the absence of reflux during aspiration and was confirmed by opacification of the central venous device with contrast medium. During opacification, all central venous devices with an obstructed fibrin sheath displayed the same contrast medium flow pattern: part of the contrast medium flowed backward from the distal hole of the catheter, along the outside wall of the catheter, and inside the fibrin sheath until exiting into the venous circulation upstream from the tip of the catheter with some of the contrast medium remaining in the sheath.
Two patients with a guidewire that was inadvertently embolized during placement of a central venous device were immediately referred for percutaneous extraction of the guidewire. In both cases, the distal tip of the guidewire was located in the right atrium.
Venous access was achieved via the femoral route in 152 of the 156 venous procedures in this study. The jugular route or brachial route was used only in the three patients who had iliac or inferior vena cava thrombosis and in one patient who had an inferior vena cava filter. All procedures were performed on an outpatient basis with the patient under local anesthesia except for 11 patients who had already been admitted to the hospital and the eight patients who were younger than 10 years and thus required general anesthesic for the procedure.
Retrieval of the embolized catheters and the two guidewires was undertaken with loop snares. These snares were most often custom-made using a 7-French catheter (FSF catheter; Cook Europe, Bjaeverskov, Denmark) with a 3-m-long 0.018-inch guidewire (Angiodyn Fuhrungsdraht; B. Braun, Melsungen, Germany) folded in two or a 9-French catheter (FSF catheter; Cook Europe) with a 3-m-long 0.021-inch guidewire (Angiodyn Fuhrungsdraht; B. Braun). However, in pediatric patients and in patients with a central venous device that had to be snared in the heart or the pulmonary artery, a ready-to-use nitinol snare measuring 15-30 mm in diameter (Amplatz goose neck and Snare catheter; Microvena, White Bear Lake, MN) was usually preferred. The catheter was retrieved via the puncture site without the need for an introducer sheath in most of the patients. The two guidewires were retrieved using a femoral approach after grasping each guidewire as close as possible to the tip with a nitinol snare. The unfolded guidewires were then extracted with a 10-French introducer.
If initially located in the heart or the pulmonary artery, an embolized catheter was sometimes repositioned in the vena cava before snaring. In these cases, a 6-French angled pigtail catheter (Torcon Avantage angiography angled pigtail catheter; Cook Europe) was placed alongside the embolized catheter and was repeatedly rotated until it enfolded the embolized catheter. The pigtail catheter was then gently withdrawn, and the embolized catheter was repositioned in a location where it would be easy to snare, most often in the inferior vena cava. The catheter was then released by rotating the pigtail catheter in the opposite direction.
Flexible triple grasping forceps (Captura three-prong grasper; Cook Europe), balloon catheters, and wire baskets (Dormia; Cook Europe) were used only if difficulties in retrieving the device were encountered using these techniques.
Replacing a catheter with a distal tip that had migrated was always achieved with a 5- or 6-French pigtail catheter (Torcon Avantage angiography angled pigtail catheter; Cook Europe) using the same technique as that described for repositioning of embolized catheters before snaring.
Fibrin-sheath stripping was performed using a nitinol loop snare. The snare was inserted in the femoral vein and opened beyond the tip of the catheter. The snare was then advanced to encircle the distal tip of the malfunctioning catheter. The snare was closed around the proximal part of the catheter, and mild to moderate tension was exerted while the snare was slowly withdrawn. This maneuver was repeated until contrast medium injected through the port flowed normally through the distal tip of the catheter.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
We failed to retrieve five embolized central venous devices because the catheter lacked a free end (n = 3) or had embolized distally in small vessels (n = 2). The three catheters that lacked a free end were located in the superior vena cava, exactly where they were initially positioned; however, these catheters had split in the costoclavicular space and could not be dislodged with a pigtail catheter or grasping forceps or grasped with loop snares. The two distally embolized catheters were short fragments of a catheter located distally in a small branch of the pulmonary artery (n = 1) or in a thymic vein (n = 1). Despite attempts to mobilize catheter fragments with grasping forceps, balloons, and wire baskets, retrieving these catheters was not possible, probably because of endothelialization.
Migrated catheters and distal tips were successfully repositioned in 32 of 38 attempts (Fig. 2A,2B). However, only 24 of the 32 replaced central venous devices could be used thereafter because the remaining eight were too short. Seven of the 24 catheters that underwent tip replacement migrated again during the 3-month follow-up. These seven catheters were then retrieved and exchanged for new ones if central venous access was still needed. Replacement was therefore effective in only 17 of the 38 migrated catheters. The six repositioning failures occurred for catheters that had been inserted deeply into small veins (vertebral, thyroid, mammary) and could not be dislodged.
|
|
Fibrin-sheath stripping (Fig. 3A,3B,3C,3D) was always successful, and the central venous device could be reused at least during the subsequent 3 months of follow-up. A longer follow-up period was not possible because all these patients died from cancer or related complications.
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Given these risks, embolized catheters should be removed in most instances, and percutaneous endovascular retrieval should be the technique of choice when possible. Embolized catheters should be removed because the procedure is feasible, safe, and efficient, as shown in this study; this study is, to our knowledge, the largest series of extractions of central venous device catheter fragments since the first report about the endovascular removal of a foreign body in 1964 [5].
Experience has taught us that repositioning catheters that are difficult to snare in their initial location with a pigtail catheter before retrieval is attempted is a useful maneuver that facilitates snaring and retrieval considerably. Repositioning often facilitates the removal of a catheter that lacks a free distal tip. The absence of a free end has been reported by most authors to be the main cause of failure [2, 17, 18]. In our experience, the loop snare is the tool of choice for retrieval, and if retrieval using a loop snare fails, the likelihood of success with other tools is low. Custom-made standard loop snares are highly efficient and can be used in all instances, even if commercially available nitinol snares are increasingly used because they are easier to maneuver and therefore probably save time.
In our series, retrieval failures mainly occurred when central venous device catheters lacked a free end and were probably completely endothelialized. This situation may be the only one in which a central venous device catheter should be left in place because the risk of complication from leaving the catheter in place is probably lower than that from retrieving it. However, whether the catheter is endothelialized cannot be determined until a retrieving device is placed close to the catheter. Therefore, retrieval should be attempted in all cases.
If a catheter tip has migrated, the central venous device should not be used because of the risk of infusion into the wrong veins, which can give rise to major complications [2, 17, 18]. In cases of tip migration, we first attempted to reposition the catheter by strongly flushing the catheter, as reported elsewhere [7]. However, this technique was rarely successful in our hands, and we most often resorted to repositioning the tip with a pigtail catheter using an endovascular approach, which is easy to perform. This maneuver can be attempted even if the distal tip is not free, thereby avoiding more complex procedures such as the long loop technique [8]. However, only 17 of the 38 cases referred for catheter complications were definitively resolved, mostly because remigration often occurred because the catheter was too short. Replacing the catheter may prove pointless because it is simply too short to be used and will invariably migrate again. Therefore, the endovascular approach should be reserved for repositioning port catheters that are cumbersome to exchange, and simple catheters should be exchanged because this exchange is quite simple to perform.
Catheter stripping for fibrin-sheath obstruction is a technique that has been introduced for management of occluded hemodialysis catheters [9, 19]. Recently, long-term patency rates for hemodialysis catheters have been reported to be higher after catheter exchange than after stripping [20]. However, in our series of central venous devices in patients with cancer, the stripping technique restored patency in 16 of 16 port catheters, with a 100% primary patency at 3 months. Although our series is small, the difference in long-term patency between hemodialysis catheters and central venous devices in patients with cancer could indicate that the causes of obstruction differ between the two types of catheter. The so-called fibrin sheath of the dialysis catheter has been shown to be associated with thrombus [21]. Because the technique of stripping is less invasive to perform than port replacement, the number of stripping interventions will probably increase in the future because patency can be maintained in a simple and highly effective way.
For several years, interventional radiologists have assumed the primary role in all central venous access procedures, including placement of chest ports, because of safety, efficacy, and cost-effectiveness. In our experience, performing an interventional radiologic procedure to manage mechanical complications related to a central venous device was highly feasible, minimally invasive, safe, and highly effective. The procedures were neither technically difficult nor did they require sophisticated devices. These procedures can thus be performed in any angiography suite with standard angiography tools, such as a pigtail catheter and a custom-made loop snare. We were able to retrieve most of the embolized catheters and eliminate the fibrin-sheath obstruction. However, results were less satisfactory for repositioning the catheter after tip migration, mainly because the catheter was too short.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Sobolevsky, R. G. Sheiman, S. Faintuch, and L. Perry Femoral-to-Port Through-and-Through Wire Access to Reestablish Subcutaneous Port Function Am. J. Roentgenol., April 1, 2007; 188(4): 1047 - 1049. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dhanani, S. Senthuran, R. Olivotto, R. J. Boots, and J. Lipman The entrapped central venous catheter Br. J. Anaesth., January 1, 2007; 98(1): 89 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. J. Chwals Vascular Access for Home Intravenous Therapy in Children JPEN J Parenter Enteral Nutr, January 1, 2006; 30(1_suppl): S65 - S69. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Murphy, A. E. Healey, J. Harper, S. Saltissi, and R. G. McWilliams Percutaneous removal of a right atrial catheter fragment: the value of the En Snare Nephrol. Dial. Transplant., October 1, 2004; 19(10): 2686 - 2687. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
