| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology, the University of Chicago Hospitals, 5841 S Maryland Ave, MC 2026, Chicago, IL 60637 (B.F., J.A.L., J.N.L., T.V.H., J.D.R.); and the Racine Radiology Group, Wis (G.X.Z.). Received March 13, 2000; revision requested May 2; revision received May 30; accepted June 28. Address correspondence to B.F. (e-mail: bfunaki@midway.uchicago.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: A femoral venous approach was used to recanalize occluded veins or catheterize small collateral veins in 24 patients in whom all major central veins were occluded. A loop snare or catheter was used as a target for antegrade puncture. Metallic stents were deployed if necessary. Once antegrade access was secured, catheters were placed in a conventional fashion.
RESULTS: Technical success was achieved in 22 (88%) of 25 procedures (one patient underwent two procedures). All catheters functioned immediately after placement. There were two procedural complications: a vasovagal episode requiring intravenously administered atropine sulfate and an episode of respiratory distress requiring intubation. There were no instances of pneumothorax, nerve injury, or bleeding complications. Catheter malfunction requiring exchange occurred at a rate of 0.67 per 100 catheter days. Infection requiring catheter removal occurred at a rate of 0.06 per 100 catheter days. Primary patency was 90% at 1 month, 71% at 6 months, and 25% at 12 months. Secondary patency was 100% at 6 months and 70% at 12 months.
CONCLUSION: In patients undergoing hemodialysis in whom conventional venous access sites have been exhausted, interventional radiologic venous recanalization for the placement of permanent catheters is safe and effective. Catheters placed in recanalized veins or small collateral veins have shorter primary patency rates compared with those of conventionally placed catheters, but the former can be maintained for relatively long periods.
Index terms: Catheters and catheterization, central venous access, 907.1269, 9462.1269 • Catheters and catheterization, complications, 907.442, 9462.442 • Catheters and catheterization, technology • Dialysis, 81.42 • Veins, access, 907.1269, 9462.1269
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Procedure
Written informed consent was obtained prior to all procedures after the nature of the procedure had been explained. All patients were given a choice between attempted recanalization and femoral or inferior vena cava catheter placement. The chest, neck, and right groin were cleansed in standard fashion with a povidone iodine scrub (Clinidine solution; Clinipad, Rocky Hill, Conn). Conscious sedation with fentanyl citrate (Sublimaze; Abbott Laboratories, North Chicago, Ill) and midazolam hydrochloride (Versed; Roche Pharmaceuticals, Manati, PR) was provided. All patients received intravenously administered ceftizoxime sodium (Cefizox; Fujisawa Healthcare, Deerfield, Ill; 1 g) as prophylaxis.
Standard sterile technique was used. The right common femoral vein was punctured by using the Seldinger technique, and an 80-cm 9-F vascular introducer sheath (Cook, Bloomington, Ind), or an 80-cm 8-F sheath (Super Arrow-flex percutaneous sheath; Arrow International, Reading, Pa) was inserted and advanced into the superior vena cava. A 180-cm rigid hydrophilic guide wire (Glidewire; Boston Scientific, Natick, Mass) and a hydrophilic coated 5-F catheter (Angled Taper Glidecath; Boston Scientific) were advanced into the brachiocephalic veins. The catheter and guide wire were used to engage any vein in the upper chest or neck. Neck veins were chosen preferentially. Catheters were placed in subclavian veins only in patients who were not candidates for an upper extremity fistula or graft. Up to 30 minutes were used in an attempt to catheterize a suitable vein.
We modified our procedure slightly as we gained experience. Once an appropriate vein was catheterized, US was performed in all later cases (n = 10) to choose a suitable puncture site away from overlying arteries. The catheter was used as a target for puncture. Access into neck veins was obtained by using a micropuncture set (Cook), which consisted of a 21-gauge needle, a 0.018-inch guide wire, and coaxial 3- and 5-F dilators. If the vein was successfully accessed (n = 7), a guide wire was advanced into the superior vena cava. If access was initially unsuccessful, the catheter was exchanged for a 10-mm snare (Microvena, White Bear Lake, Minn) that was opened in the vessel and used as a target (n = 15). With fluoroscopic guidance, the loop portion was punctured by using a 21-gauge needle (micropuncture set), the needle was exchanged for a 0.018-inch guide wire, and the snare was closed and used to pull the guide wire into the superior vena cava.
In later cases (n = 10), the coaxial dilators included in the micropuncture set were advanced over the 0.018-inch guide wire, and the inner dilator was removed. A side-arm adapter (Passage hemostasis valve; Merit Medical, Galway, Ireland) was attached to the outer dilator, and iodixanol (Visipaque; Nycomed Amersham, Princeton, NJ; 320 mg of iodine per milliliter) was slowly injected while the dilator was retracted over the 0.018-inch guide wire through the subcutaneous tissues to ensure that through-and-through arterial puncture had not occurred. Once this was excluded, the guide wire was exchanged for a 0.035-inch wire (Superstiff Amplatz wire; Boston Scientific), and a dual-lumen hemodialysis catheter was placed in standard fashion.
If the superior vena cava was occluded, angioplasty was performed. If a stenosis of greater than 50% persisted after angioplasty, an appropriately sized Wallstent (Boston Scientific) was deployed and dilated with an appropriately sized angioplasty balloon catheter (Blue Max; Boston Scientific). In two patients, metallic stents (Wallstent; Boston Scientific) were deployed in a recanalized external jugular vein and in an occluded subclavian vein to facilitate dilator-sheath insertion. After catheter insertion, the femoral venous sheath was removed, and hemostasis was achieved. Procedural time varied from 45 to 90 minutes.
Follow-up
The vein used for catheter insertion, complications, catheter types, and number of times catheter dysfunction and infection occurred were recorded. All patients except one were followed up in the nephrology department and were referred back to the radiology department for follow-up catheter intervention. One patient was followed up by his referring physician at another hospital. Primary patency was defined as the time from placement until any catheter intervention in our department. Secondary patency was defined as the time from placement until the site was abandoned, regardless of the number of interventions. Kaplan-Meier analysis was performed.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Recanalization was unsuccessful in three patients. In these patients, there was complete occlusion of the jugular veins, subclavian veins, brachiocephalic veins and superior vena cavae. No suitable veins were catheterized despite prolonged attempts with catheters and guide wires. Superior vena cava occlusions were crossed and dilated by using angioplasty balloon catheters in seven patients. Suboptimal results after angioplasty necessitated metallic stent placement in three patients. Two additional stents were placed in a thrombosed subclavian vein and an external jugular vein to aid in catheter insertion. All stents were successfully inserted, and no stent-related complications occurred during follow-up.
Two procedural complications occurred. A 77-year-old woman had a vasovagal reaction and became bradycardic and hypotensive (heart rate, 45 beats per minute; blood pressure, 90 mm Hg systolic and 50 mm Hg diastolic) during attempts to recanalize her occluded right internal jugular vein. The procedure was temporarily stopped, and 1 mg of atropine sulfate (Abbott Laboratories) was administered intravenously. Her heart rate and blood pressure rapidly returned to baseline, and the remainder of the procedural and postprocedural period was uneventful.
A 42-year-old man became short of breath approximately 15 minutes after successful catheter insertion. His oxygen saturation decreased to 85% despite his receiving oxygen with a rebreather mask. Both naloxone hydrochloride (Narcan; Astra Pharmaceutical Products, Westborough, Mass; 1 mg) and flumazenil (Romazicon; Hoffman-La Roche, Nutley, NJ; 0.2 mg) were administered intravenously, without improvement. Fluoroscopy was performed in the chest and revealed no evidence of a pneumothorax or pleural effusion. Visual inspection revealed no evidence of a neck hematoma. The anesthesiology staff performed urgent intubation, and the patient’s condition immediately stabilized. Pulmonary arteriography was performed and revealed no evidence of a pulmonary embolus. The patient was transferred to the intensive care unit in stable condition and made an uneventful recovery. A presumed diagnosis of aspiration was made, although the exact cause of the patient’s distress remains unconfirmed at this time. His catheter continued to function well 1 month after the procedure.
There were no cases of pneumothorax, hemothorax, inadvertent arterial puncture, hematoma, nerve injury, or catheter malposition.
Follow-up
Total follow-up duration was 6,234 days. One patient was lost to follow-up. By using life table analysis, primary patency was 90% at 1 month, 71% at 6 months, and 25% at 12 months (Fig 1). Secondary patency was 100% at 6 months and 70% at 12 months (Fig 2).
|
|
In three patients, catheters were removed, and permanent femoral catheters were inserted. One patient later underwent a second successful recanalization procedure. In a 65-year-old woman with multiple medical problems, the catheter was exchanged over a guide wire, and the patient recovered after a course of antibiotics. Infection requiring catheter removal or exchange therefore occurred at a rate of 0.06 per 100 catheter days. Two catheters were inadvertently removed at 8 and 9 weeks in a 62-year-old woman and a 44-year-old man. The woman’s catheter was replaced by means of fluoroscopy-guided puncture into the lumen of a deployed Wallstent. The man delayed seeking medical attention for several days and was admitted to another hospital. Multiple failed attempts at central venous catheterization occurred prior to transfer to our hospital. He died due to lack of dialysis prior to the attempted catheter replacement in our department.
Three additional patients died during follow-up. None of the deaths were related to catheter placement or use. A 34-year-old man with a psychiatric disorder intentionally pulled out his catheter 2 weeks after insertion. It was replaced by using the same Wallstent puncture technique described previously.
In four patients, catheters were removed after successful upper extremity synthetic arteriovenous graft creation 37, 50, 123, and 172 days after catheter insertion.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
In terms of catheter function, the path to a patent central vein is unimportant. As long as the catheter tip resides in a large central vein or in the right atrium, adequate flow rates for hemodialysis can be obtained. Our experience supports the safety and effectiveness of venous recanalization in patients in whom conventional access sites have been exhausted, as reported previously (1,2). Catheters inserted with this technique have a slightly higher incidence of dysfunction compared with that of conventional internal jugular venous catheters, but they have a lower incidence compared with that of femoral venous catheters (5,6).
The right internal jugular vein is the preferred access site for the placement of hemodialysis catheters. The subclavian veins are generally avoided because of the risk of central venous stenoses, which may render the ipsilateral extremity unfit for shunt or fistula creation. Trerotola and associates (5) reported an infection rate of 0.08 per 100 catheter days and a malfunction rate of 0.22 per 100 catheter days with right internal jugular vein catheters.
If the internal and external jugular veins become occluded, other routes of access are used, such as the femoral vein and inferior vena cava. Zaleski and associates (6) reported their experience with 41 tunneled femoral venous hemodialysis catheters in 21 patients. An infection rate of 0.24 per 100 access days was observed. Primary patency at 6 and 12 months was 78% and 55%, respectively. Secondary patency at 6 and 12 months was 95% and 61%. Lund et al (7) reported a series of 17 inferior vena cava catheters used for hemodialysis. The rate of thrombosis was 0.33 per 100 access days, and the rate of infection was 0.28 per 100 access days. In addition, two catheters developed defects that required removal. Primary patency was 52% at 6 months and 17% at 12 months.
Our results with recanalization compare favorably with these published results. Catheter malfunction requiring exchange occurred at a rate of 0.67 per 100 catheter days. Infection requiring catheter removal occurred at a rate of 0.06 per 100 catheter days. Primary patency was 71% at 6 months and 25% at 12 months. Secondary patency was 100% at 6 months and 70% at 12 months. Of course, recanalization has an inherent advantage compared with the use of other alternative access sites, since it preserves other sites for future use.
In general, we have observed two types of central venous occlusions. Complete venous thrombosis is commonly seen after infection and is recognized sonographically prior to venipuncture. Focal central venous occlusions may also occur with patent proximal and distal segments. This type of obstruction is usually discovered after venipuncture when difficulty is encountered in the passing of a guide wire into the central veins.
In patients with patent jugular veins and a more central occlusion, we first attempted to cross the occlusion by using an antegrade approach, since this shorter and straighter approach provides greater mechanical advantage compared with a retrograde (ie, femoral) approach. Although absence of respiratory variation may indicate a central venous occlusion (8), the existence of a central venous obstruction may be clinically silent and is often unknown prior to attempted catheter insertion.
When an antegrade approach is unsuccessful, either because of complete central venous thrombosis or because a central occlusion cannot be crossed from above, retrograde recanalization is used. This approach is advantageous when catheters are deployed into thyrocervical collateral veins, since many of these veins are either discontinuous or too tortuous to support a large dual-lumen hemodialysis catheter. A retrograde approach allows a relatively straight and smooth course to be chosen. An additional strength of the retrograde technique is that once through-and-through access has been secured, the large diameter hemodialysis catheter can be deployed in veins even smaller than the catheter itself (Fig 3). In fact, insertion into small collateral veins may be preferable, since these veins are unlikely to contribute substantially to venous return.
|
|
|
|
|
We have found that the type of chronic complications associated with catheters placed in small collateral or recanalized veins are identical to those encountered with conventional catheters (ie, dysfunction due to fibrin sheath formation, inadvertent removal, and infection). Our treatment approach for these problems was likewise identical for conventional catheters. When dysfunction occurred, guide-wire exchange, occasionally in conjunction with balloon dilation, was used; it was proved effective in disrupting fibrin sheaths and restoring adequate flow rates. Fibrin stripping with the use of snares would also likely be as effective as standard catheters, since the catheter tip is located in the low superior vena cava or in the high right atrium similar to conventional catheters. Suspected catheter infection initially was treated with systemic antibiotics. If symptoms did not resolve, catheters were removed or exchanged.
Recanalization of occluded veins may be a tedious and time-consuming procedure (1). A thorough knowledge of venous collateral drainage and adjacent structures is important. Access sites must be carefully chosen to avoid arterial or brachial plexus injury. Whenever possible, we generally selected anterior superficial veins away from the brachial plexus. Our technique evolved slightly as we gained experience. In all later cases, two steps were taken to avoid inadvertent arterial puncture. Prior to venipuncture in the neck, US was performed to identify and avoid superficial arteries. After access was achieved, contrast medium was injected during fluoroscopy into a side-arm adapter over a 0.018-inch guide wire while the catheter was slowly retracted through the subcutaneous tissues. In this manner, opacification of a small artery would indicate an inadvertent through-and-through arterial puncture.
All of our patients had undergone numerous catheter insertions in the past, and most had extensive scar tissue in the neck and upper chest. In this situation, catheter insertion can be difficult even with through-and-through guide-wire access. The scar tissue crimps and may even kink the peel-away sheath used for catheter deployment, precluding advancement of the catheter into the central veins. Therefore, in two early patients, balloon dilation and stents were used to enlarge veins to facilitate catheter insertion. Stents are useful to fashion and maintain a favorable course for catheters and to help prevent catheter kinking. They are also extremely helpful in patients with superior vena cava obstruction. In two patients, stents were used as targets for repuncture and allowed rapid catheter replacement after inadvertent catheter removal. Although none of our patients suffered a stent infection, this remains a concern and potential drawback of stent insertion, especially since patients with hemodialysis catheters are prone to developing catheter infections. Therefore, stents are deployed only when they are deemed absolutely necessary.
All of the failures in our series occurred due to our inability to recanalize occluded vessels or catheterize suitable thyrocervical collateral veins. Using the snare technique, we were always able to successfully insert catheters in any vein that we could successfully catheterize with a retrograde approach. We have found that the snare can be formed, even in occluded segments of veins. We believe either that the loop opens between the chronic thrombus and the vessel wall or that catheterization attempts create a cavity large enough for formation of the loop.
Recently, Horton and associates (3) reported a slightly different method for hemodialysis catheter insertion into occluded veins. In their technique, a patent arm or neck vein is punctured, and a sheath is advanced across occluded central veins. The sheath is then used as a target for puncture and as a portal to advance a guide wire into the superior vena cava. The disadvantage of this technique is that after central venous access is achieved, the sheath must be snared from the femoral vein, cut off at the hub, and pulled out of the femoral vein so as not to disrupt chest wall access. In addition, the technique may fail despite successful guide-wire traversal due to an inability to advance a sheath through the occluded segment. In contrast, we were always able to place a catheter when occlusions were successfully traversed. Moreover, we used the retrograde technique only after antegrade placement had already failed.
In summary, we have found that venous recanalization is an effective method for permanent hemodialysis catheter insertion. Although catheters inserted in this manner have a slightly higher incidence of dysfunction compared with that of conventional catheters, they are well tolerated by patients and preserve limited venous access sites in patients with end-stage renal disease.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Funaki Central Venous Access: A Primer for the Diagnostic Radiologist Am. J. Roentgenol., August 1, 2002; 179(2): 309 - 318. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
