| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Departments of Radiology (S.P., J.M.L., M.W.W., S.R.M., R.K.K., R.L.G.), Surgery (J.M.P.), and Medicine (H.L.S., H.E.I., R.S.O., M.G., B.R.D.), University of California San Francisco, Box 0628, 505 Parnassus Ave, San Francisco, CA 94143-0628. From the 1998 RSNA scientific assembly. Received September 16, 1998; revision requested November 19; final revision received July 2, 1999; accepted July 30. Address reprint requests to S.P. (e-mail: perini@itsa.ucsf.edu).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: A prospective study of the Tesio catheter was performed. Follow-up data regarding catheter function and adequacy of dialysis were obtained from nine hemodialysis facilities.
RESULTS: Seventy-nine Tesio catheters were placed in 71 patients. Immediate technical success was 99% (78 of 79 catheters). The procedure complication rate was 9% (seven catheters). Only two complications required intervention: one fatal air embolism and one chest wall hematoma. Sixty-seven catheters in 60 patients were followed up for a total of 4,367 catheter days. Overall, catheter-related infection occurred in 9% (six of 67 catheters). Primary catheter patency was 87% at 1 week, 82% at 1 month, 72% at 3 months, and 66% at 6 months. Mean blood flow was 286 mL/min immediately after insertion, 301 mL/min at 3 months, and 306 mL/min at 6 months. Adequate dialysis dose as reflected by a urea reduction ratio of 60 or more or a urea kinetic modeling, or Kt/V, value of 1.2 or more was observed on at least one occasion for 74% and 76% of catheters, respectively.
CONCLUSION: The Tesio catheter is a reasonable means of vascular access for patients who undergo dialysis but are not candidates for surgical shunt placement.
Index terms: Catheters and catheterization, central venous access, 907.1269 • Catheters and catheterization, complications, 81.42, 907.1269 • Catheters and catheterization, technology, 907.1269 • Dialysis, 81.42 • Interventional procedures, 907.1269
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Tunneled dialysis catheters are now commonly inserted by interventional radiologists. The technical aspects of catheter insertion have been evaluated for a variety of catheters, but to our knowledge, the performance characteristics of these catheters have not been extensively studied (2,3). Of particular interest is the adequacy of dialysis that can be achieved with these catheters.
At our institution, requests for tunneled dialysis catheters have increased steadily over the past few years, and we have chosen to use the Tesio twin dialysis catheter system (Medcomp; Harleysville, Pa) as our primary tunneled dialysis catheter. The Tesio system consists of two separate 10-F silicone catheters that are placed percutaneously with a Seldinger technique and tunneled in the subcutaneous tissues to an exit site on the chest. Theoretically, the twin catheter design offers the possibility of superior performance by providing a large cross-sectional diameter and allowing variable separation of arterial and venous catheter tips (4–6).
To assess the performance of the Tesio catheter, we conducted a nonrandomized, prospective, observational trial evaluating its use as a means of vascular access for hemodialysis. Evaluated outcomes included the technical success of radiologic insertion, procedure complications, catheter patency, and infection rates. Catheter function was assessed in accordance with the definitions of catheter malfunction and primary failure introduced by the Dialysis Outcomes Quality Initiative (DOQI) of the National Kidney Foundation (1). In addition, adequacy of delivered dialysis was determined with standard methods.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Subject Enrollment
All patients referred for Tesio catheter placement during a 9-month study period (August 1997 to April 1998) were considered for enrollment. Seventy-nine Tesio catheter placement procedures were performed in 71 patients. Six patients underwent Tesio catheter placement during the study interval but were not enrolled because they were unable or unwilling to sign informed consent. Ten patients enrolled in part 1 of the study were excluded from part 2 because they indicated they would be obtaining maintenance dialysis at a facility outside of our area.
The demographic data for the 71 enrolled patients (41 female and 30 male patients; mean age, 60.7 years; age range, 12–98 years) are shown in Table 1. Two minor patients, aged 12 and 15 years, were included in the study. Sixty-five patients underwent placement of one Tesio twin catheter pair. Five patients underwent two sequential Tesio catheter placement procedures. One patient required serial placement of four catheter pairs. Results were compiled and analyzed for each Tesio twin catheter pair.
|
Tesio Twin Catheter System
The Tesio twin catheter system comprises two 10-F catheters that are designed for insertion through the internal jugular vein. The catheters exit the skin along the chest wall medial and superior to the breast. The venous and arterial catheter tips are positioned in the right atrium and at the junction of the right atrium and superior vena cava, respectively. A football-shaped Dacron retention cuff is attached to the catheter. The catheter exit site is chosen so that the cuff is positioned just proximal to the skin exit site. Catheters may be inserted through the right or left internal jugular vein or through other venous access when necessary.
Catheter Insertion
Catheter insertion was performed by one of six attending physicians (J.M.L., J.M.P., M.W.W., S.R.M., R.K.K., R.L.G.) experienced in placement of the Tesio catheter system. The primary operator was assisted by one of three interventional radiology fellows (S.P. and others) or a second attending physician. Procedures were performed in the interventional radiology suite. C-arm fluoroscopy was used in 77 of 79 cases. Patients were placed in Trendelenburg position during the procedure whenever a tilting table was used (54 cases).
Procedures were performed with conscious sedation with intravenously administered sedatives. General anesthesia with airway intubation was used in two of 79 cases. Prophylactic antibiotics were administered in seven cases.
Color Doppler ultrasonography (US) was performed to locate the internal jugular vein, verify patency, and mark an appropriate puncture site in the lower neck within the Sedlow triangle. In women, breast tissue was retracted inferiorly with waterproof tape. The neck and chest were scrubbed with chlorhexadine gluconate (Hibiclens; Zeneca Pharmaceutical, Wilmington, Del) and draped in a sterile fashion. Local anesthesia was obtained at the venous puncture site by subcutaneously injecting 1% mepivacaine hydrochloride (Polocaine-MPF; Astra USA, Westborough, Mass). The internal jugular vein was accessed with a 21-gauge micropuncture set (Cook, Bloomington, Ind) with real-time US guidance, and a 0.038-inch J-shaped flexible wire was advanced into the superior vena cava. A second 0.038-inch J-shaped flexible wire was then advanced into the superior vena cava. This wire was introduced through the initial venous puncture site after insertion of a 6-F sheath in 60 of 79 cases (76%). A second puncture with the 21-gauge needle was used at a site 1–2 cm superior to the initial puncture site in 19 cases (24%). A 1–2-cm-long vertical incision was then made at the insertion site with a number 11 blade.
The twin Tesio catheters were flushed with saline solution containing heparin and fitted with homeostatic plastic clamps. An 11-F peel-away sheath was placed over the inferior guide wire. The venous Tesio catheter was then advanced though the peel-away sheath and positioned with its tip in the right atrium. A second 11-F peel-away sheath was then placed over the remaining guide wire, and the arterial Tesio catheter was introduced through the sheath and its tip positioned in the superior vena cava 3 cm above the tip of the venous catheter. Manual compression was applied at the venous insertion site, and tunneling along the chest wall was performed.
In 55 (70%) of 79 cases, formation of the catheter tunnels was facilitated by creating a 2-cm-long horizontal incision (referred to as a "counter incision") 1–2 cm inferior to the clavicle. Blunt dissection was performed with a Schnidt clamp to create a subcutaneous tunnel from the counter incision, superficial to the clavicle, and back to the puncture site in the internal jugular vein. The two catheters were sequentially pulled under the skin from the neck incision to the subclavicular counter incision with the Schnidt clamp.
Catheter exit sites were chosen and marked 7–10 cm below the clavicle, medial and superior to the breast, and several centimeters apart. Local anesthesia along the length of the planned subcutaneous tunnel and at the selected exit site was obtained by means of infiltration with mepivacaine hydrochloride. For each catheter, a subcutaneous tunnel and skin exit site were created with the sharp trocar from the Tesio catheter kit. The tunnel was dilated to the skin exit site with the football-shaped dilator attached to the threaded end of the tunneling trocar. After the dilator was removed, the catheter was attached to the threaded end of the trocar and pulled through the tunnel and out the exit site.
The catheter hubs were attached, and the catheters were flushed with saline solution containing heparin. The position of each catheter was verified fluoroscopically. The neck and counter incisions were closed with a 4.0 Dexon II (David and Geck, Danbury, Conn) running subcuticular stitch. Vertical 0-silk mattress sutures were used at the venous insertion site when bleeding and made placement of a subcuticular stitch difficult. Skin closures (Steri-Strips; Medical-Surgical Division/3M, St Paul, Minn), sterile gauze, and plastic adhesive dressings were applied to the wounds. An upright chest radiograph was obtained at the end of the procedure. Patients were instructed to leave the sterile dressings in place for 2 days and keep the wounds dry for 5–7 days at which time the skin closures could be removed.
Follow-up Interventions
Catheter stripping was performed as an outpatient procedure with standard techniques. One Tesio catheter pair was replaced by exchanging each catheter over a wire after the catheters were dissected free near the previous venous puncture site. New subcutaneous tunnels were created in this case.
Poor blood flow during dialysis was treated by administering urokinase (Abbott Laboratories, North Chicago, Ill), 5,000 U diluted in a volume to fill the catheter lumen, followed by aspiration after 30 minutes.
Procedure Data Collection
Data forms were completed at the time of each insertion procedure. Patient demographic data, indication for catheter placement, prior hemodialysis history, and planned outpatient dialysis center were recorded. Any medical history of diabetes, congestive heart failure, or a coagulopathy was noted.
Coagulation parameters were recorded including prothrombin time, partial thromboplastin time, and platelet count. A platelet count of greater than or equal to 140 x 109/L and a prothrombin time less than 14.7 seconds (international normalized ratio, <1.5) were considered normal. Insertion site, number of venous puncture sites, use of counter incision, use of preprocedure antibiotics or blood products, procedure duration, technical success, and immediate complications were also recorded.
Dialysis
Dialysis was performed at one of nine dialysis centers: two inpatient facilities and seven free-standing outpatient dialysis facilities. Different dialysis machines were used (models 2008D, 2008E, and 2008H, Fresenius USA, Walnut Creek, Calif; models C2 and C3, Cobe BCT, Wheat Ridge, Colo). In general, high-flux dialyzers were used in the outpatient setting, and conventional dialyzers were used for hospitalized patients.
Follow-up Data Collection
Data concerning catheter performance and adequacy of delivered dialysis were collected from each facility providing dialysis to enrolled patients during frequent site visits. All dialysis session flow sheets, nursing notes, orders, and laboratory values were reviewed. Nursing and support staffs at each facility were interviewed regarding catheter function and problems in the interval since the last site visit. Any instances of poor blood flow during dialysis, treatment with urokinase either during or at completion of dialysis, early termination of dialysis related to either poor blood flow or aberrant venous or arterial pressures were recorded.
Pump blood flow, prescribed blood flow, time of dialysis, venous pressure, and arterial pressure were recorded. Data points were obtained for the first dialysis and subsequent dialysis sessions closest to 1 week and 1, 3, and 6 months after catheter placement. For pump blood flow, venous pressure, and arterial pressure, the maximum, minimum, and mean values during a dialysis session were recorded. For patients new to hemodialysis or patients who tolerated dialysis at high blood flow rates poorly, low blood flow rates were prescribed. To differentiate prescribed low blood flow from that related to poor access function, mean pump blood flow was compared with the prescribed value, and a percentage of the prescribed flow was calculated for each dialysis session. Mean blood flow and mean percentage of prescribed blood flow were determined for the initial, 1-week, and 1-, 3-, and 6-month time points. Since dialysis sessions did not always coincide with the preselected time points, a mean catheter day for the analyzed subgroup was calculated at each time point and used to report mean blood flow. High venous pressure (>350 mmHg), low negative arterial pressure (<350 mmHg), or "sucking" in the arterial circuit were the parameters used to indicate access capacity was being exceeded. Pump blood flow rates were reduced if these conditions occurred.
Adequacy of dialysis was assessed on the basis of the URR and Kt/V. URR is calculated as 1 - R, where R is the ratio of plasma urea level before and after dialysis. The expression Kt/V is used to quantify prescribed and delivered dialysis. In the expression, K is the dialyzer clearance (in milliliters per minute), t is the dialysis duration (minutes), and V is the urea volume (in milliliters) for the patient. Kt/V, therefore, equals the fraction of the patient's urea volume cleared during dialysis. Calculation of Kt/V directly with this formula yields the prescribed dialysis dose. Kt/V is used to determine the actual dose of dialysis delivered. In the simplest single-compartment model, a patient's urea volume is assumed to include both intra- and extracellular water. If a fixed volume is assumed and residual renal function, weight change, and urea generation during dialysis are ignored, Kt/V varies approximately with the natural logarithm of the fractional reduction in blood urea nitrogen (ie, Kt/V = loge[C1/C2], where C1 is the predialysis and C2 the postdialysis blood urea nitrogen levels). In current practice, a dialysis prescription is usually designed to achieve a URR of 60 or more and Kt/V of 1.2 or more.
On the basis of the routine protocol at our outpatient dialysis facilities, the URR and Kt/V were obtained on a monthly basis. Therefore, the interval between catheter placement and initial calculations of URR and Kt/V varied but was approximately 1 month or less. Values closest to the preselected values at the initial, 1-week, and 1-, 3-, and 6-month time points were used. As with the blood flow data, the mean catheter day for the collected values was calculated as a time point for reporting the mean Kt/V and URR.
The recirculation percentage was not routinely determined at any of the chronic-care facilities at which enrolled patients received dialysis. For a dialysis session of normal duration, delivery of adequate dialysis, as indicated by an expected URR or Kt/V excludes clinically important recirculation. The routine calculation of the recirculation percentage was not added to the prescription for care of patients enrolled in the study. No recirculation percentages were calculated for enrolled patients.
At the time of catheter removal, a data sheet was completed by the interventional radiologist to indicate the reason for the removal and any procedure complications.
Statistical Analysis
All data were maintained in a database created with a computer program (ACCESS; Microsoft, Redmond, Wash). A primary catheter failure rate was determined according to DOQI guidelines (1). Primary and primary-assisted catheter patency were calculated with the Kaplan-Meier product limit method. The end point of primary patency was poor function resulting in catheter removal, over-the-wire catheter exchange, or intracatheter treatment with urokinase. The end point of primary-assisted patency was poor function resulting in catheter removal or over-the-wire catheter exchange. Patient death, patient loss to follow-up, catheter removal secondary to availability of alternative access, catheter removal after improvement in renal function, and catheter removal secondary to infection were considered censoring events for both primary and primary-assisted patency.
The rate of catheter removal secondary to malfunction was calculated as a function of catheter days and as a percentage of total catheters placed. Rates for total catheter infections and infections requiring catheter removal were determined. The mean time of infection with respect to the date of catheter placement was also calculated.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
Procedure Complications
There were seven complications, for an overall complication rate of 9% (seven of 79 cases). Intervention was required in two of these cases, defining a rate of 2% (two cases) for complications requiring intervention. The first complication requiring intervention was a large neck and chest hematoma that necessitated transfusion with two units of packed red blood cells. The second was a fatal air embolism. This patient had experienced a myocardial infarction 5 days before the procedure and was tachypnic at the time of the procedure. Air embolism resulting in hypotension occurred during insertion of the 11-F sheath. The patient was immediately placed in a left decubitus position with an initial improvement in hemodynamics. Fifteen minutes later, he sustained an acute cardiac arrest and could not be successfully resuscitated.
There were five complications that did not require intervention: one pneumothorax, two neck hematomas, one hemothorax, and one case of prolonged oozing at the catheter exit site. There were no episodes of inadvertent carotid artery puncture, cardiac arrhythmia requiring intervention, hematoma requiring evacuation, or pneumothorax requiring chest tube placement.
Part 2: Patient Outcomes and Duration of Follow-up
In part 2 of this study, 67 Tesio catheter pairs in 60 patients were followed up during dialysis (Table 3). Ten of the patients enrolled in part 1 of the study were excluded from part 2 because the patients were not planning to have dialysis at a facility in our area. One patient died of complications related to his brain tumor after successful catheter placement but before dialysis was initiated. Of the 67 catheter pairs that were followed up, 35 were patent at the end of the study period, 19 functioning catheters were removed, eight were lost to follow-up because the patients moved to a dialysis facility outside of our area, and five were censored from follow-up owing to patient death for reasons unrelated to the presence of the Tesio catheter. The causes of death in these five patients included myocardial infarction (n = 2), cerebral hemorrhage (n = 2), and multiorgan system failure (n = 1). For the patients who moved to a remote dialysis facility during the study, data were recorded until the time of transfer. Patients were followed up for a mean 65 days (range, 1–236 days) and a total of 4,367 catheter days. More than 6 months of follow-up data were available for seven catheter pairs.
|
Overall, treatment with urokinase was performed 24 times in 14 different catheter pairs. Intervention for poor function was, therefore, necessary in 14 (21%) of 67 cases or at a rate of 0.55 per 100 catheter days. Early termination of dialysis because of poor catheter function occurred 12 times in nine (13%) of 67 catheters, defining a rate of 0.27 per 100 catheter days.
With the Kaplan-Meier product limit method, primary catheter patency was 87% at 1 week, 82% at 1 month, 72% at 3 months, and 66% at 6 months (Fig 1a). Primary-assisted catheter patency was 100% at 1 week, 96% at 1 month, and 92% at 3 and 6 months (Fig 1b).
|
|
Catheter-related Infection
Overall, catheter-related infection occurred in 9% (six of 67) of cases (six catheter pairs in four patients). These six instances define a rate of 0.14 per 100 catheter days for catheter-related infection. The infections occurred a mean 45 days after catheter placement (range, 19–104 days). Two of these infections occurred in a patient with positive results for human immunodeficiency virus and a CD4 count of 90. Two others occurred in a patient who was receiving immunosuppression therapy because of cardiac transplantation. Four of the six infections (four catheter pairs in three patients) necessitated catheter removal. The four instances define a rate of 0.09 per 100 catheter days for infection requiring catheter removal. Symptomatic new central venous thrombosis was not observed in any patient during the follow-up period.
Catheter Performance
Catheter dysfunction is defined by DOQI guidelines 23 and 34 as an inability to achieve a blood flow rate of at least 300 mL/min. In dialysis sessions in which the prescribed flow rate was less than 300 mL/min, inability to achieve the prescribed flow rate was also considered catheter dysfunction. With this definition, catheter dysfunction was seen in 34% (20 of 59), 28% (16 of 58), 39% (20 of 51), 38% (eight of 21), and 38% (three of eight) of catheters at the immediate, 1-week, and 1-, 3-, and 6-month time points, respectively.
Blood flow data for dialysis performed within 3 days of catheter placement was available for 55 of 67 catheters. Initial use of the Tesio catheter was delayed beyond this time point in the remaining 12 cases. The mean blood flow rate at the immediate time point was 286 mL/min (range, 160–400 mL/min). Blood flow rates less than 300 mL/min were prescribed for 22% (12 of 55) of patients during the initial dialysis session. Prescriptions less than 200 mL/min were given to 4% (two of 55) of patients. One of these patients underwent dialysis at 160 mL/min. The mean observed blood flow was 92% of the mean prescribed blood flow at the immediate time point.
At time points of 1 week (range, 5–9 days; 55 catheters), 1 month (range, 26–33 days; 45 catheters), 3 months (range, 84–98 days; 18 catheters), and 6 months (range, 181–184 days; seven catheters), mean blood flow rates of 307 mL/min (range, 198–420 mL/min), 304 mL/min (range, 200–430 mL/min), 301 mL/min (range, 208–430 mL/min), and 306 mL/min (range, 250–350 mL/min), respectively, were seen (Fig 2). Blood flow rates less than 300 mL/min were prescribed for 7% (four of 55), 4% (two of 45), and 11% (two of 18) of patients at the 1-week and 1- and 3-month time points, respectively. No patients had a prescribed blood flow less than 300 mL/min at 6 months. At the 1-week to 6-month time points, the mean observed blood flow rates reflected 85%–89% of the mean prescribed rates (Fig 3).
|
|
URR or Kt/V were available at one or more time points for 43 catheters. The mean time for dialysis in patients in this group was 2.9 hours. These data were not obtained for 24 catheters that were in place 35 days or less (mean, 15 days; range, 4–35 days) because the assessment was scheduled monthly. Actually, these data were obtained at a mean interval of 42 days (range, 18–90 days). The URR was measured 90 times, and the overall mean was 60.5. Kt/V was calculated 75 times, and the overall mean was 1.24.
At the initial time point, a mean 2 days after catheter placement (range, 0–4 days), the mean URR for catheters in the analyzed subgroup was 56.3. At time points of 1 week (mean, 11 days; range, 6–15 days), 1 month (mean, 34 days; range, 17–56 days), 3 months (mean, 83 days; range, 60–112 days), and 6 months (mean, 171 days; range, 143–193 days), the mean URR was 62.2, 60.6, 60.3, and 65.3, respectively. Mean Kt/V values of 1.05, 1.36, 1.29, 1.24, and 1.31 were determined at the same time points. The mean duration of dialysis in hours at these time points was 2.62, 3.00, 3.00, 2.84, and 3.03, respectively.
A URR of 60 or more was measured during at least one dialysis session for 31 of 42 (74%) catheters. Kt/V of 1.2 or more was determined for 26 of 34 (76%) catheters on at least one occasion. At the initial, 1-week, and 1-, 3-, and 6-month time points, a URR of 60 or more or Kt/V of 1.2 or more was determined for five of 11 (45%), 20 of 29 (69%), 18 of 28 (64%), eight of 15 (53%), and seven of nine (78%) catheters, respectively.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
Because Tesio catheter placement requires two serial catheter insertions and two separate chest wall tunnels, the procedure is more difficult and time-consuming than insertion of a single-catheter, dual-lumen system. However, the mean procedure time for Tesio catheter insertion in our study was 31 minutes. It should be noted, though, that it is common practice at our teaching institution for Tesio catheter insertion procedures to be performed by two physicians, an interventional radiology attending and fellow. We believe that the expertise of two skilled operators is helpful in Tesio catheter insertion.
Despite a high success rate achieved during a reasonable procedure time, we found that Tesio catheter placement can be challenging. There are several procedure points that we believe should be emphasized. First, in patients with end-stage renal disease, bleeding may be profuse during insertion of a Tesio catheter or other tunneled catheter. The combination of high central venous pressure and poor platelet function in such patients can lead to bleeding at the jugular insertion site. Bleeding at the insertion site was usually well controlled during the procedure by means of manual compression. Bleeding at the skin exit site was treated by placing tension on the retention cuff or by using a purse string suture closure. We believe that the use of a counter incision just below the clavicle facilitates tunneling and improved hemostasis. Although this technique required an additional suture closure, it only minimally increased the overall procedure time and did not increase patient discomfort. The incision was small and overlaid the intended course of the subcutaneous tunnel, which was anesthetized. No infection was seen at the counter incision site.
Because the two catheters are independent, the Tesio twin catheter system afforded a versatility not possible with dual-lumen catheters. In patients with small central veins or partial occlusions, the two Tesio catheters could be placed separately through different veins. This capability was used in two patients in our study.
Overall, the incidence of procedure complications in our series was 9% (seven of 79 procedures). This rate is similar to that reported by Prabhu et al (7) (8% [seven of 82 catheters]). These rates are higher than that typically reported with cuffed dialysis catheters (less than 5% according to DOQI guideline 34). It is possible that compared with single-catheter, dual-lumen devices, the additional complexity of twin catheter placement may contribute to a higher complication rate. Of note, however, a complication rate of 1.0% (three of 300 catheters) was reported in a series by Tesio et al (6).
The most serious complication in our series was a fatal air embolism. Both Lund et al (2) and Trerotola et al (3) report clinically silent air embolization during the insertion of dialysis catheters. In our series, a fatal air embolism occurred in a patient who had recently sustained a myocardial infarction, had wide respiratory excursion, and was unable to cooperate. Although this complication was recognized immediately and corrective measures were instituted, a fatal outcome resulted. In retrospect, this patient may have been better treated with the less invasive placement of a nontunneled dialysis catheter. This case underscores the risks of performing this procedure in medically unstable patients or those with preexisting respiratory compromise.
Long-term Complications
Infection and malfunction are the primary factors that limit the use of catheters as a long-term means of access for dialysis. In our study, catheters were removed for infection or malfunction in seven (10%) of 67 patients.
In a prior study of Tesio catheters, Millner et al (8) reported bacteremia in four (16%) of 25 patients during a mean follow-up period of 4.3 months. They suggested the possibility that the use of two catheters may increase the risk of catheter-related infection. We did not find a higher than expected incidence of catheter infection in our series. Overall, catheter-related infection occurred in 9% (six of 67) of cases during a mean follow-up of 2.2 months. In a recent prospective trial to compare two types of dual-lumen dialysis catheters, Trerotola et al (9) reported a rate of catheter removal due to infection of 11% and 14% over a 3-month follow-up for silver-coated and non–silver-coated catheters, respectively. In another recent study (10), a risk of catheter infection as high as 40% was reported during a 9-month follow-up period of 16,081 catheter days. These data suggest the rate of infection with the Tesio catheter is similar to that reported with other tunneled dialysis catheters.
The other major limitation of tunneled catheters is malfunction that manifests as an inability to deliver adequate extracorporeal blood flow rate or as catheter thrombosis. Catheters were removed in three patients for this reason, defining a removal rate of 0.069 episodes per 100 catheter days at risk. This malfunction rate is lower than that previously published for tunneled dual-lumen single catheters. In a study with the Bard double-lumen dialysis catheter, Lund et al (2) reported a catheter removal rate of 0.39 episodes per 100 catheter days at risk (63 catheters removed during a follow-up period of 16,240 catheter days).
Central to any discussion of catheter removal for malfunction are the threshold parameters used to guide clinicians in the identification and removal of malfunctioning catheters. The recent DOQI guidelines impose a flow rate definition of less than 300 mL/min for catheter malfunction. Herein, we report our results in accordance with this standard. It is important to note, however, that our referring nephrologists do not necessarily use this threshold value in determining catheter malfunction in actual practice. Rather, they typically use a threshold value of 200 mL/min in combination with other machine-related parameters and measures of dialysis adequacy as described herein. This explains why our primary catheter failure rate according to DOQI guidelines (40%) was higher than our rate of catheter removal for malfunction (4%).
Catheter Performance and Adequacy of Delivered Dialysis
In current practice, the two most widely accepted measures of delivered dialysis are URR and Kt/V. In general, target values of a URR of 60 or more and a Kt/V of 1.2 or more indicate adequate dialysis.
To our knowledge, surprisingly few data are available for catheter access systems regarding the adequacy of delivered dialysis as defined on the basis of URR or Kt/V. Atherikul et al (11) report a mean Kt/V with Tesio catheters of 1.39. This compared with mean values of 1.42 and 1.32 for the PermCath (Quinton Instrument, Seattle, Wash) and the VasCath, respectively. The mean Kt/V values in our study were somewhat lower than that reported by Atherikul et al. However, at time points beyond the initial dialysis, the mean values were all 1.2 or more (range, 1.24–1.36). Overall, we found that targeted values of adequacy were achieved in about 75% of our patients (URR 
60 in 74%, Kt/V 1.2 in 76%).
Catheter flow rate is an important parameter that may limit the quantity of dialysis that can be delivered. However, it is important to recognize that delivered dialysis is dependent on a number of factors in addition to blood flow rate, including time for dialysis, machine-specific factors (dialysate flow rate, membrane type, membrane surface area, and dialysate composition), and patient urea volume. Time for dialysis is prescribed by the treating nephrologist but may also be affected by patient compliance and tolerance. For patients new to hemodialysis or those intolerant of dialysis at high blood flow rates, low blood flow rates (<300 mL/min) may be prescribed.
In general, with current dialysis equipment, blood flow rates of 300–400 mL/min are required to provide adequate dialysis and achieve target levels of a URR of 60 or more and a Kt/V of 1.2 or more during a 3-hour dialysis session. Blood flow rates of 350–400 mL/min are routinely achieved with surgical arteriovenous access. In our study, mean blood flow rates with the Tesio catheter consistently averaged about 300 mL/min and did not vary appreciably during a 6-month period (range, 301–307 mL/min). It is important to recognize that although the mean blood flow was 300 mL/min or more, 28%–39% of patients at each time point were treated at flow rates less than 300 mL/min. Low blood flow rates, therefore, likely contributed to the inadequate delivery of dialysis that was observed in about 25% of patients.
The relatively low flow rates reported in our study may be due to factors other than catheter design. For example, a uniform and aggressive approach to the maximizing of catheter blood flow rates was not possible in our study because patients were treated at nine different dialysis centers under the care of a variety of nephrologists. One of our patients, for example, underwent dialysis at 200 mL/min throughout the entire interval of enrollment, and a flow rate less than 300 mL/min was prescribed in 16 (24%) of 67 patients for their first dialysis session. Furthermore, the majority of our patients underwent dialysis routinely at a prescribed rate between 300 and 350 mL/min despite the fact that the catheters may have been able to support a higher flow rate. Nevertheless, the reported mean flow rates represent 85%–92% of the mean prescribed flow rates. We believe that the mean blood flow rates recorded in this study reflect what can be reasonably achieved in the outpatient chronic dialysis setting.
In the three prior studies shown in Table 4 that included the Tesio catheter, mean flow rates varied from 276 to 400 mL/min. In the study by Tesio et al (6), a blood flow rate less than 200 mL/min was seen in only 5% of dialysis treatments over a period of 1,420 patient-months. In the study by Prabhu et al (7), 95% of patients had a flow rate greater than 375 mL/min. In a recent controlled study by Atherikul et al (11), mean blood flow rate during 30 consecutive sessions with the Tesio catheter in 22 patients was 396 mL/min. This is compared with blood flow rates of 437 mL/min for surgical access, 384 mL/min with the PermCath, and 320 mL/min with the VasCath.
In conclusion, at our institution, the Tesio catheter provides an important alternative means of vascular access for patients who are not candidates for surgical methods. Radiologically guided placement of the Tesio catheter device can be accomplished safely in most cases; however, given our fatal complication and the relative complexity of the placement procedure, we recommend reservation of this procedure for medically stable patients who are not in distress. Together, catheter-related infection and low blood flow necessitated catheter removal in 10% of patients, which is comparable to data reported with other catheters. In the majority of patients, targeted measures of dialysis adequacy can be achieved. A substantial fraction of patients, however, failed to receive adequate dialysis at each interrogated time point. The mean blood flow rate of approximately 300 mL/min in our study is less than that reported by some prior investigators and likely contributed to inadequate dialysis in some patients.
We conclude that the Tesio catheter provides an acceptable but not optimal means of providing vascular access for hemodialysis.
Definitions
Catheter dysfunction.—Inability to deliver extracorporeal blood flow rate of at least 300 mL/min. In dialysis sessions in which the prescribed flow rate was less than 300 mL/min, inability to achieve the prescribed flow rate was considered catheter dysfunction.
Catheter-related infection.—Infection related to the presence of the catheter including the following: (a) exit site infection (redness, swelling, or purulent drainage within 3 cm of the catheter site); (b) tunnel infection (redness, swelling, or purulent drainage more than 3 cm from the exit site and along the course of the subcutaneous tunnel); and (c) catheter-related bacteremia (positive blood cultures with or without positive catheter tip culture). Patients with fever and a negative blood culture were not considered to have a catheter infection.
Delayed complications.—Complications occurring remote from the catheter placement procedure including infection, catheter malfunction, and venous thrombosis.
Primary failure rate.—The fraction of catheters unable to deliver adequate blood flow of 300 mL/min or more during the first attempted dialysis session.
Procedure complications.—Immediate complications evident during the procedure including cardiac arrhythmia, pneumothorax, air embolism, and bleeding.
Procedure time.—Time from needle puncture to application of dressing.
Technical success.—Ability to successfully insert the twin Tesio catheters and position them appropriately.
|
Footnotes |
|---|
Author contributions: Guarantors of integrity of entire study, S.P., J.M.L.; study concepts, all authors; study design, S.P., J.M.L., J.M.P., H.L.S., M.W.W., R.K.K., R.L.G.; definition of intellectual content, S.P., J.M.L., J.M.P., H.L.S., M.W.W., R.K.K., R.L.G.; literature research, S.P., J.M.L.; clinical studies, S.P., J.M.L., J.M.P., M.W.W., S.R.M., R.K.K., R.L.G.; data acquisition, S.P.; data analysis, S.P., J.M.L.; manuscript preparation, S.P., J.M.L.; manuscript editing and review, all authors.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. J. Oliver, D. C. Mendelssohn, R. R. Quinn, E. P. Richardson, D. K. Rajan, R. A. Pugash, J. A. Hiller, A. J. Kiss, and C. E. Lok Catheter Patency and Function after Catheter Sheath Disruption: A Pilot Study Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1201 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wang, J. M. LaBerge, G. M. Chertow, R. K. Kerlan, M. W. Wilson, and R. L. Gordon Tesio Catheter Access for Long-term Maintenance Hemodialysis Radiology, October 1, 2006; 241(1): 284 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Coronel, J. A. Herrero, P. Mateos, M. L. Illescas, J. Torrente, and M. J. d. Valle Long-term experience with the Thomas shunt, the forgotten permanent vascular access for haemodialysis Nephrol. Dial. Transplant., September 1, 2001; 16(9): 1845 - 1849. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
