| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Diagnostic Imaging, Foothills Medical Centre, 1403 29th St., N.W., Calgary, Alberta T2N 2T9, Canada.
Received August 2, 2001;
accepted after revision January 29, 2002.
Address correspondence to J. K. Wong.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
7 days) of these catheters in our practice. SUBJECTS AND METHODS. Data were prospectively collected for 639 radiologically placed tunneled hemodialysis catheters. The reason for catheter removal was recorded in each case. Tips of removed catheters were routinely sent for microbial culture.
RESULTS. Fifty-two (8.1%) of 639 catheters were removed within 7 days of insertion. Six (0.9%) of these had completed their purpose and had not failed; these were not included in the study. Of the 46 catheters having early failure, six (0.9%) were clotted and 12 (1.9%) were suspected of being infected, only three of which had a proven catheter-related infection. Twenty-eight catheters (4.4%) were removed for other reasons. In this group, the most common reasons were poor tip position (n = 9) and catheter replacement over a guidewire into a preexisting fibrin sheath (n = 8). Only two failed because of poor tip orientation. Other reasons for failure were kinked or pinched catheters (n = 4) and bleeding (n = 3), including one exsanguination, and two unknown reasons.
CONCLUSION. By paying careful attention to catheter tip position, searching diligently for the presence of a fibrin sheath when catheter exchanges are made over a wire, and better investigating presumed catheter infection, we could reduce the early failure rate by more than half, from 46 cases to 20 cases (nine cases of suspected infection were in fact not infected).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Common indications for tunneled central venous catheters include antibiotic therapy, chemotherapy, parenteral nutrition, access for fluids, and hemodialysis. The use of long-term tunneled catheters for hemodialysis is an alternative to arteriovenous fistula or polytetrafluoroethylene grafts in selected kidney patients. This catheter access is unfortunately a necessary long-term option in patients in whom all other surgical access has been exhausted. However, the more common scenario is the use of central venous catheters as a bridge to a maturing arterio-venous fistula or graft or to a renal transplant. In our service, the most common indication for central venous access is hemodialysis; approximately 33% of all hemodialysis patients use a central venous catheter at any given time. Indications for placement of central venous catheters for hemodialysis include an arteriovenous fistula or graft maturing, a failed site, acute renal failure, and as definitive hemodialysis access in selected patients. Central venous catheter placement for hemodialysis is different from placement for other indications because high flow rates are needed for hemodialysis, which necessitates proper orientation and placement of the catheter. Thus, these catheters are more susceptible to failure.
The purpose of our study is to analyze the causes of early failure in tunneled hemodialysis catheters. We present the immediate complication rates, defined as catheter removal on or before day 7 after catheter placement, of 639 consecutively placed tunneled hemodialysis catheters.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patency of the vein dictated which vein was cannulated. The right internal jugular vein was the primary site of catheter placement, the left internal jugular vein was the second site, and the subclavian or femoral veins were the remaining sites. In patients with occluded internal jugular veins, the subclavian or femoral veins were chosen, depending on the patient's activity level and on whether the subclavian veins needed to be preserved for future arteriovenous fistula or polytetrafluoroethylene graft formation.
All procedures were performed in the angiography suite after informed consent was obtained. A cap, mask, and surgical gown were worn by both the interventionalist and the nurse. They also donned sterile gloves after doing a complete surgical scrub. The patient also wore a cap. The surgical site was scrubbed three times with betadine. At the discretion of the radiologist, patients were given 2-5 mg of diazepam IV (Valium; Roche, Mississauga, Ontario, Canada) or 1-2 mg of midazolam hydrochloride (Versed; Roche) and 50-100 µg of fentanyl citrate IV (Faulding; Dorval, Quebec, Canada). However, most patients required no IV sedation. Lidocaine hydrochloride 1% was used as a local anesthetic at the site of venous cannulation and at the tunnel in all patients unless contraindicated. The target vein was punctured under sonographic guidance (7.5-MHz transducer SSD-2000; Aloka, Tokyo, Japan) with a 19-gauge needle. A 0.035-inch guidewire was then passed into the distal superior vena cava and secured to the drapes. A chest wall tunnel was created using standard techniques as previously described [4]. Either a Gamcath (Gambro, Toronto, Ontario, Canada) or a Vascath (Vas-cath, Mississauga, Ontario, Canada) catheter was positioned in the upper right atrium if possible, although 1 cm proximal or distal to the superior vena cava and the right atrial junction was also considered an acceptable position. Once a catheter was placed, fluoroscopy was used to assess tip position, kinking, and appropriate orientation of the venous and arterial ports, and to monitor initial complications associated with the procedure. Ideal tip position was considered to be within 1 cm of the superior vena cava and right atrial junction as seen fluoroscopically.
All hemodialysis catheters requiring removal were referred to the radiology department. All catheters were removed under sterile conditions in the angiography suite. The reason for removal was recorded in each case. All removed catheters were routinely sent for microbiologic culture of the catheter tip. If infection was suspected, concomitant peripheral blood cultures were drawn from the patient. The catheter tip culture technique included culturing both the inner and outer surfaces of the catheter tip on blood agar. If 15 or fewer colony-forming units of bacteria grew from the catheter tip, the catheter tip was considered to be contaminated. If more than 15 colony-forming units grew from the catheter tip, the tip was considered to be infected. The results of the catheter tip culture and the blood cultures were compared. If both blood culture and catheter tip culture yielded the same organism, the patient's bacteremia was considered to be the result of an infected central venous catheter. If the patient still required hemodialysis, placement of a central venous catheter was delayed until the day of the patient's hemodialysis session, and a de novo puncture was performed if possible. However, in some patients with limited remaining access sites, guidewire exchange at the pre-existing site was performed. During guidewire exchange, the existing catheter was accessed at the neck entry site, and a guidewire was passed into the superior vena cava. The old catheter was then cut and removed over the guidewire. A new sheath was placed over the guidewire, through which a new catheter was placed, and a new tunnel to the venotomy was created.
During the study, no catheter-related venous thrombosis or fibrin sheath formation was treated with thrombolytic therapy because our hospital policy mandates a minimum 24-hr course in the ICU despite the dose of thrombolytic agent used.
In this study, early failure was defined as catheter removal on or before day 7 after the initial catheter placement. Follow-up until the line was removed was obtained in all patients.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
An early failure of catheter function was defined as removal of the catheter within 7 days of initial placement, with the exception of catheters that had completed their purpose within 7 days; these catheters were not included in the early failure group. Of the 639 central venous catheters placed, only 52 were removed within 7 days of placement. Six of the 52 catheters removed within 7 days were removed because they fulfilled their purpose. Hence, 46 catheters in this study were considered to have malfunctioned early. Table 2 shows the reasons for early failure. Of the 46 catheters having early failure, 12 catheters (2%) had suspected infection and six (1%) were removed because of occlusion. Twenty-eight catheters (4%) were removed for other reasons (all with poor clinical dialysis function), with the two largest categories being poor positioning (n = 9) and fibrin sheath formation (n = 8).
|
One case of early failure occurred when the tip of the catheter was pressed against the atrium (Fig. 1). This catheter failed at 7 days after placement. A second case of early failure occurred in a large woman with pendulous breasts in whom placement was technically difficult (Fig. 2A,2B). The initial position of this catheter was in the proximal right atrium (Fig. 2A), but subsequent migration of the catheter tip to the level of the junction of the innominate vein and the superior vena cava caused failure (Fig. 2B). Another cause of early failure was a catheter twisting on its long axis, which resulted in poor function (Fig. 3). However, the tip of this catheter was in good position. Orientation of a catheter was also a factor in early failure. In one case, a hemodialysis catheter malfunctioned with low arterial pressures on the first dialysis attempt because the arterial port was abutting the right atrial wall (Fig. 4A,4B). Dialysis was successful after the catheter was manipulated so that the arterial port was not adjacent to the wall. A kinked central venous catheter at the neck entry site resulted in early failure when the catheter malfunctioned, resulting in poor flow on the first hemodialysis session (Fig. 5).
|
|
|
|
|
|
|
Of the 12 catheters removed for suspected infection, only three (0.5%) had a proven catheter-related bacteremia. No lines were removed because of a tunnel infection.
Seven of the eight early failures caused by fibrin sheath formation occurred during guidewire exchanges. This fact was subsequently confirmed when the patient returned to the radiology department after a hemodialysis session that took place after the catheter exchange over a guidewire. Another guidewire exchange was performed; however, contrast material was injected through a dilator to assess the presence of a fibrin sheath (fibrin sheaths were documented in seven of eight cases). Figure 6A shows the presence of a fibrin sheath where contrast material was injected through a sheath during guidewire exchange of a tunneled dialysis catheter. Balloon angioplasty was performed, and subsequent contrast injection shows disruption of the fibrin sheath (Fig. 6B).
|
|
One early failure was the result of exsanguination when the patient pulled the catheter out at home.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Central venous catheters are invaluable for vascular access for hemodialysis. Traditionally, hemodialysis catheters are used temporarily as a bridging device until an arteriovenous fistula has matured or placement of a graft has been completed. Central venous catheters are also significantly used as the definitive mode for vascular access and hemodialysis in a subset of dialysis patients who have no other viable alternative. At our center, 33% of the dialysis patients use a central venous catheter as the definitive method for hemodialysis. A report by Woods et al. (presented at the American Society of Nephrology meeting, November 1996) shows a trend toward the increasing use of central venous catheters in patients with chronic renal disease. The incidence of central venous catheter use for hemodialysis has increased to the point that 40% of the total hemodialysis population uses a catheter, compared with only 15% of patients using catheters for hemodialysis in 1991 [9]. Hemodialysis catheters are unique compared with other catheters in that high flow rates are required; optimal function requires flow rates of greater than 250 mL/min through both the arterial and venous ports. Slight changes in position, thrombus formation, kinking, and poor orientation have greater adverse effects on hemodialysis catheter function than on nonhemodialysis catheters in which high flow rates are not required.
Studies have shown that infection is a major complication of central venous catheters and usually necessitates their removal [10, 11]. In our study, the largest single group of early failures was those caused by clinically suspected infection of the indwelling catheter. Twelve catheters suspected of being infected were removed, with only three catheters yielding positive bacterial culture results consistent with the bacterial blood cultures. Thus, most catheters (75%) in this early failure group did not have a proven infection by our assay, suggesting that nine of 12 catheters may not have been the source of the bacteremia and did not need to be removed. No catheters were removed for clinically suspected tunnel infection. Catheters were removed only if catheter tip infection was suspected. A promising study by Capdevila et al. [11] showed a success rate of 100% with systemic antibiotics in eradicating 13 episodes of catheter-related bacteremia without necessitating catheter removal. In contrast, Marr et al. [12] showed a high incidence of bacteremia (40%) in their hemodialysis population with poor catheter salvage when antibiotic therapy alone was used. Only 32% of the catheters in patients with proven bacteremia were salvaged with antibiotic therapy as the sole treatment. Other studies have also shown similar results [7, 13], suggesting that antibiotic therapy alone is inadequate for catheter salvage and elimination of bacteremia. However, Marr et al. also showed that the rate of complications does not increase with attempted catheter salvage. Furthermore, successful eradication of bacteremia without catheter removal is more likely to occur in patients infected with Gram-positive organisms [7, 12]. Our results show an unnecessarily high rate of catheter removal for presumed catheter-related infections; we suggest that antimicrobial therapy be initiated while awaiting blood culture results. This procedure will likely decrease the number of catheters removed because of presumed catheter-related bacteremia. Because the catheter placement is performed with a sterile technique and cuffed tunneled catheters are used, the early failure of the catheter caused by catheter-related infection is unlikely. If the patient does indeed have bacteremia, a trial of antibiotics may be appropriate to attempt to salvage the catheter; however, the patient's clinical status will dictate treatment. Additionally, preservation of the site using a guidewire to exchange a new catheter for an infected catheter does not increase the infection rate [14,15,16,17]. In fact, catheter exchange over a guidewire with empiric antibiotic therapy has been shown to be an excellent choice in dealing with catheter-associated bacteremia [16, 17]. Thus, we suggest a trial of antibiotic therapy first and, if that fails, then catheter exchange over a guidewire with empiric antibiotic therapy.
Our next largest causes of early failure in catheters were poor positioning and poor orientation. The current practice of catheter placement is to initially place the catheter tip at the junction of the superior vena cava and the right atrium or in the proximal right atrium [10]. An important factor in catheter placement is the significant migration of the catheter when the patient moves from a supine to an upright position [18, 19]. Patients who are obese and women with pendulous breasts are more likely to have significant catheter tip migration [19, 20]. Both Nazarian et al. [19] and Kowalski et al. [18] suggest initially placing the catheter farther into the right atrium to account for the cephalad migration of the catheter when the patient is standing. Nazarian et al. also found that catheters placed via the jugular route moved less than those placed via the subclavian route, which is another good theoretic reason for the internal jugular vein approach. Therefore, early failure can be minimized with better initial placement of the catheter, with the tip being placed farther into the right atrium.
In our study we actually had one patient in whom the catheter migrated farther into the right atrium, instead of the usual cephalad migration, so that the catheter was eventually lying against the floor of the right atrium (Fig. 1). This catheter failed by the seventh day after placement. One possible explanation for the caudal migration of this catheter is severe volume overload; once the patient underwent hemodialysis and lost the extra volume, the unusual and unexpected caudal migration occurred.
In hemodialysis, catheter placement must also take into account the arterial port, which cannot lie adjacent to the vessel wall if the catheter is to function optimally. Figure 4A,4B shows the arterial port lying against the venous wall; this catheter subsequently failed on the first attempt at hemodialysis. Manipulation of the catheter by introducing a guidewire into each port and twisting the catheter so that the arterial port was not abutting the vessel wall restored proper function and a high flow rate. In some cases, twisting will not remove the arterial port from its position against the wall, because the catheter can be twisted only so much before it breaks. Thus, at the time of placement it is imperative to closely examine the orientation of the tip to ensure the arterial port is not adjacent to the vessel wall and that prompt aspiration of at least 5 mL/sec per port is possible. At our center, we routinely have the venous port cephalad during insertion through the peelaway sheath; however, this practice occasionally still results in having the arterial port abut the vessel wall.
Occasionally, catheter placement may result in the catheter twisting and causing malfunction. Twisting may cause an improper arterial port position as just described, or it may cause a kink in the catheter, which will cause malfunction. Twisting may also be significant enough that it causes the lumen of the catheter to narrow, although that is rare. In Figure 3, initial placement of the catheter caused twisting on its long axis and resulted in an early failure. Proper tunneling and minimal twisting during catheter placement should prevent this from happening.
Kinking generally occurs in very thin patients with little subcutaneous fat and a relatively steep apical curve that results from a narrow chest diameter. The curve can be widened in these patients by creating a two-step tunnel, or by using a more lateral approach to the internal jugular vein instead of an anterolateral approach. Kinks can be cleared by bluntly dissecting a large space around the apex of the curve and then advancing the catheter over a wire, or withdrawing it slightly if length and cuff position allow.
Fibrin sheath formation and central venous catheter clotting have been well described as causes of catheter malfunction. In our study, fibrin sheath formation accounted for eight (1.3% of all hemodialysis catheters placed) of 46 failed catheters, and clotting accounted for six (0.9%) of 46 failed central venous catheters. A widely accepted method of treating suspected fibrin sheath formation or clot is to use a thrombolytic agent to biochemically disrupt the fibrin sheath [21, 22]. Mechanical disruption has also been tried, with varying success [23, 24]. In our center, when we perform a guidewire exchange we now search diligently for the presence of a fibrin sheath. If a fibrin sheath is found, then an angioplasty balloon is inflated at the site of the fibrin sheath to mechanically disrupt it. Our preliminary results are highly favorable using angioplasty of a fibrin sheath and subsequent guidewire catheter replacement; this procedure has patency rates comparable to those of de novo catheter placement (study in progress, data not presented).
In conclusion, hemodialysis central venous catheters are an integral part of treatment in the patient with renal failure. In our center, hemodialysis catheter placement is a major component of our interventional radiology practice, with hemodialysis via a catheter being performed in 33% of all hemodialysis patients. By ensuring proper catheter tip position, proper catheter orientation, and a smooth arc to prevent kinking, and by detecting fibrin sheaths and subsequently disrupting them mechanically, we could easily halve our early failure rate. Furthermore, better investigation of presumed catheter-related infection would also improve our early failure rate, because only one quarter of catheters removed had a proven catheter-related bacteremia. Adhering to these procedures will not only improve patient care by ensuring optimal hemodialysis function but will also improve the efficiency of a busy interventional radiology practice.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Falk Use of the brachiocephalic vein for placement of tunneled hemodialysis catheters. Am. J. Roentgenol., September 1, 2006; 187(3): 773 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Smith, J. M. Ryan, and D. N. Reddan Transhepatic Catheter Access for Hemodialysis Radiology, July 1, 2004; 232(1): 246 - 251. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
