- Email: [email protected]
Complication Rates Observed in Silicone and Polyurethane Catheters of Totally Implanted Central Venous Access Devices Implanted in the Upper Arm
CLINICAL STUDY
Complication Rates Observed in Silicone and Polyurethane Catheters of Totally Implanted Central Venous Access Devices Implanted in the Upper Arm Jasmin D. Busch, MD, Maren Vens, PhD, Catherine Mahler, MD, Jochen Herrmann, MD, Gerhard Adam, MD, and Harald Ittrich, MD
ABSTRACT Purpose: To present frequency and types of complications related to silicone (SI) versus polyurethane (PUR) catheters of totally implanted venous access devices (TIVADs) placed in the upper arm. Material and Methods: A cohort of 2,491 consecutive patients with TIVADs implanted between 2006 and 2015 was retrospectively analyzed. Complications were classified according to SIR guidelines. Pearson c2 test was used for categorical variables, and Student t test was used for continuous variables. Nominal P values were reported, and 2-sided P values < .05 were considered significant. Results: Of 2,270 patients meeting the inclusion criteria, 538 had an SI catheter, and 1,732 had a PUR catheter. Total dwell time was 584,853 catheter days. Mean total complication rate was 12.25% (SI, 14.87%; PUR, 11.43%; P ¼ .040). Subanalysis revealed significant differences for material failures (eg, catheter fracture [SI, 3.35%; PUR, 0.06%; P < .001] and thrombotic catheter occlusion/venous thromboses [SI, 2.79%/0.74%; PUR, 1.33%/3.17%; P < .001]) but nonsignificant differences for infections (eg, local infection and catheter-related sepsis [SI, 4.64%; PUR, 4.68%; P ¼ 1]) or other nonthrombotic dysfunctions (eg, catheter detachment, line migration, wound dehiscence [SI, 3.35%; PUR, 2.19%; P ¼ .179]). Conclusions: The reported data suggest different risk profiles in SI catheters compared with PUR catheters, with more material failures and thrombotic catheter occlusions in SI catheters and more venous thromboses in PUR catheters.
ABBREVIATIONS PUR ¼ polyurethane, SI ¼ silicone, TIVAD ¼ totally implanted venous access device
Reliable and convenient vascular access is an integral component of modern multimodality treatment, including chemotherapy, medication administration, parenteral nutrition, supportive treatment, and blood products. Totally implanted venous access devices (TIVADs) such as port systems have gained wide acceptance with increased
From the Section of Pediatric Radiology (J.D.B., J.H.), Department of Diagnostic and Interventional Radiology and Nuclear Medicine (G.A., H.I.), Department of Medical Biometry and Epidemiology (M.V.), and Department of Legal Medicine (C.M.), University Medical Center Hamburg-Eppendorf, Martinistrasse 52, Hamburg 20246, Germany. Received February 16, 2017; final revision received April 16, 2017; accepted April 24, 2017. Address correspondence to J.D.B.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–7 http://dx.doi.org/10.1016/j.jvir.2017.04.024
availability of catheters of different materials, mainly silicone (SI) or polyurethane (PUR) (1,2). Regardless of the benefit (3), in vivo TIVADs are exposed to multifactorial strain and are associated with complications, such as infection, thrombosis, or material failure (2,4–6). Although most complications are non–life-threatening events, they may lead to interruption of medical treatment. Essentially 2 types of TIVADs have been used since initiating percutaneous placement in our department in 2006 (Fig 1a, b). Despite a similar design of multicomponent devices used, there are distinct differences regarding the attached catheters. The Cook Vital-Port Mini Titanium (Cook, Inc, Bloomington, Indiana) is attached to an SI catheter, and the Titanium SlimPort (Bard Access Systems, Inc, Salt Lake City, Utah) is attached to a PUR catheter. The purpose of this study was to present the frequency and types of complications related to an SI catheter versus a PUR catheter for TIVAD placed in the upper arm over a 10-year period in a retrospective single-center observation.
2 ▪ Complication Rates of Central Venous Access Catheters
Busch et al ▪ JVIR
Figure 1. Differently designed TIVADs. (a) Cook Vital-Port Mini Titanium composed of a circular, tapered port hub (arrowhead) attached to a 5.0-F SI catheter and (b) Titanium SlimPort with a trapezoid port hub (arrowhead) attached to a 6.0-F polyurethane catheter.
MATERIALS AND METHODS Study Design The local institutional review board approved this retrospective single-center cohort study and waived the requirement for written informed consent. Between January 2006 and December 2015, 2,491 patients were referred to the interventional radiology department for fluoroscopic-guided implantation of a TIVAD in the upper arm. Data pertaining to the first device for each patient were analyzed. Further inclusion criteria were patients 18 years old at time of procedure, completely documented implantation of either a Cook Vital-Port Mini Titanium or a Titanium SlimPort, and documented follow-up time interval > 1 day. Patient data were censored at the time of first complication, device removal, date of last follow-up, or death. The end of the study was March 2016 (3 months after last TIVAD placement).
TIVAD Implantation Technique Hemostatic disorders (ie, platelet count < 50,000/μL and international normalized ratio > 1.5), bacteremia, and septicemia were regarded as absolute contraindications. Typically, the nondominant arm was used, with the exception of thrombosis of the implantation site, ipsilateral axillary lymph node dissection, or previously embedded pacemakers or automated implantable cardioverter-defibrillators. The basilic vein was the preferred and prevailing access site, but in individual cases of insurmountable puncture obstacles (eg, preexisting thrombosis), accessible veins were used alternatively. All procedures were performed by a radiologist with at least 2 years of practice in venous access procedures. Before implantation, deep vein thrombosis of the accessed side was evaluated for using venography (Integris Allura; Philips Healthcare, Best, Netherlands). Thereafter, under strictly aseptic conditions and after local anesthesia, contrast venography-guided vein access was achieved. Port hub placement was performed in the lower third of the upper arm, and the catheter line was tunneled from the puncture
site to the port pocket. In accordance with current recommendations, patients did not receive prophylactic antibiotics or routine anticoagulation during the follow-up period. After successful insertion with fluoroscopic confirmation of correct catheter tip positioning at the cavoatrial junction (defined as 2 vertebral bodies below the carina) (7), implanted systems were flushed with sterile heparinized saline (5 mL of a solution containing 100 IU/mL) (3,8).
Specification of TIVADs From January 2006 to February 2011, the Cook Vital-Port Mini Titanium was used in 553 consecutive patients. This device comprises a tapered titanium port hub (base radius 19.0 mm, height 7.2 mm) attached to a single-lumen 5.0-F SI catheter with inner/outer diameter of 1.0/1.7 mm. From January 2011 to December 2015, the Titanium SlimPort was used in 1,938 consecutive patients. This device comprises a tapered trapezoid titanium port hub (base 19.0 24 mm, height 9.4 mm) attached to a single-lumen 6.0-F PUR catheter with inner/outer diameter of 1.3/2.0 mm.
Follow-up Patients were assessed by an interventional radiologist 1–4 days after implantation and were immediately referred to the interventional radiology department if TIVAD-related complications were suspected. Medical data were obtained retrospectively by reviewing the radiologic information system (Centricity; GE Healthcare, Little Chalfont, United Kingdom). In patients who were lost to follow-up before removal of the TIVAD or who died within the study period, the last date of documented regular catheter usage was recorded.
Data Analysis The following characteristics were assessed: technical feasibility, successful implantation with correct catheter tip positioning and aspiration of blood as well as flushing of the
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
entire system with saline without any difficulty; dwell time, calculated from the day of placement until removal (in days) or, in patients who unable to be monitored, the last date of documented regular catheter usage; and complications, classified according to the Society of Interventional Radiology (SIR) guidelines (9). Events were additionally divided into early onset ( 30 d after implantation) or late onset (> 30 d after implantation) periprocedural complications.
Statistical Analysis Analyses were conducted with Microsoft Excel 2010 (Microsoft Corp, Redmond, Washington) and R (R Foundation for Statistical Computing, Vienna, Austria). For categorical variables, Pearson c2 test with Yates continuity correction was used. Student t test was used for continuous variables. If continuous variables were assumed to be nonnormal, the Wilcoxon rank sum test was used. Nominal P values are reported without correction for multiplicity. Two-sided P values < .05 were considered significant.
Figure 2. Flow chart of the study cohort.
3
RESULTS Technical Feasibility Technical feasibility is typically not related to the catheter material itself; hence, the entire cohort was considered for evaluation. Primary implantation was technically successful with a venographically guided approach in 97.15% (2,420 of 2,491) of patients. Initial insertion failed owing to a hematoma at the puncture site (n ¼ 20), venous spasm (n ¼ 16), central vein thrombosis (n ¼ 15), failed puncture in cases of lymph edema or obesity (n ¼ 11), and per patient request during the procedure (n ¼ 9). All procedures were completed without major complications.
Patient Cohort and Follow-up The inclusion criteria were met by 2,270 patients (Fig 2) with 538 SI catheters and 1,732 PUR catheters. The total dwell time accounted for 584,853 catheter days (SI, 166,824 d; PUR, 418,029 d). In 33.44% of patients (759 of 2,270; SI,
4 ▪ Complication Rates of Central Venous Access Catheters
228; PUR, 531), the TIVAD was removed after completion of treatment. Furthermore, 21.01% of patients (477 of 2,270; SI, 185; PUR, 292) died as a result of progression of their primary disease with a functional TIVAD still in place. No death occurred secondary to device-related complications. Detailed patient information is summarized in Table 1. Characteristics of patients receiving both devices were compared to evaluate for potential bias. There were no significant differences in age distribution (SI, 59.6 y; PUR, 58.1 y; P ¼ .056) and access sites (SI, 77.0%; PUR, 72.5%; P ¼ .055). Patients who received an SI catheter were significantly more frequently male (SI, 44.8%; PUR, 38.0%; P ¼ .006). Also, dwell times were significantly longer for SI catheters (174 d; interquartile range, 59–422 d; range, 2–2,515 d) compared with PUR catheters (159 d; interquartile range, 47–318 d; range, 2–1,728 d; P ¼ .003).
Complications The total mean complication rate was 12.25% (SI, 14.87%; PUR, 11.43%; P ¼ .040), including early-onset events ( 30 d) in 3.16% of patients with SI catheters and in 3.64% of patients with PUR catheters (P ¼ .025). Regarding
Table 1. Characteristics of Patients with a CVAD between 2006 and 2015 Characteristic
SI Catheter (Cook Vital-Port Mini Titanium)
PUR Catheter (Titanium SlimPort)
Number of patients
538
1,732
Male/female
241 (44.8%)/297 (55.2%)
659 (38.0%)/1,073 (62.0%)
Age, y, mean (range) Underlying disease
59.6 (19.5–88.5)
58.1 (18.2–91.7)
Breast cancer
117 (21.7%)
451 (26.0%)
Lung cancer
111 (20.6%)
187 (10.8%)
Head and neck cancer
108 (20.1%)
274 (15.8%)
Gastrointestinal cancer
74 (13.8%)
169 (9.8%)
Gynecologic cancer
31 (5.8%)
160 (9.2%)
Lymphoma/ leukemia
31 (5.8%)
147 (8.5%)
Urogenital cancer Other malignancies
10 (1.9%) 41 (7.6%)
60 (3.5%) 228 (13.2%)
Nonmalignant disease
15 (2.8%)
56 (3.2%)
diverse types of complications, SI and PUR catheters demonstrated nonsignificant differences for infection (P ¼ 1), thrombotic dysfunction (P ¼ .395), and other nonthrombotic dysfunction (P ¼ .179) but significant differences for material failure (P < .001). Detailed information about complications is listed in Tables 2 and 3.
Infections Overall, the mean infection rate was 4.67% (SI: 4.64%, PUR: 4.68%; P ¼ 1), including 1.37% early-onset infections (SI: 1.67%, PUR: 1.27%; P ¼ .6239).
Thrombotic Dysfunction In general, thrombotic dysfunction (Fig 3) occurred with a mean rate of 4.27% in all patients, with rates of 3.53% in patients with SI catheters and 4.50% in patients with PUR catheters (P ¼ .395). However, detailed consideration revealed significant differences with catheter-induced venous thromboses (SI, 0.74%; PUR, 3.17%; P < .001) and thrombotic catheter occlusions (SI, 2.79%; PUR, 1.33%; P < .001). SI catheters had more frequent thrombotic catheter occlusions (15 vs 4 of 538 patients), whereas PUR catheters were more prone to catheter-induced venous thrombosis (55 vs 23 of 1,732 patients). For both devices, there was no significant thrombotic predisposition regarding implantation site (SI, P ¼ .269; PUR, P ¼ .437).
Nonthrombotic Dysfunction The total rate of nonthrombotic dysfunction was 3.30%. In particular, regarding material failure, the differences between the 2 devices were greatest with 3.35% of complete (n ¼ 10) and partial fractures (n ¼ 8) of SI catheters after a mean dwell time of 243 days (Figs 4 and 5a, b). In contrast, only 1 patient with a PUR catheter presented with a partial fracture at time of removal after a dwell time of 1,109 days (0.06%; P < .001). Other nonthrombotic complications were noted for both catheters with an occurrence rate of 2.47%, with nonsignificant differences of 3.35% for SI devices and 2.19% for PUR devices (P ¼ .179). Complications such as port hub rotation (SI, n ¼ 8; PUR, n ¼ 6), line migration (SI, n ¼ 6; PUR, n ¼ 15), catheter detachment (SI, n ¼ 0; PUR, n ¼ 4), effects on the median nerve (SI, n ¼ 3; PUR, n ¼ 4), wound dehiscence (SI, n ¼ 1; PUR, n ¼ 8), and delayed allergic reaction (SI, n ¼ 0; PUR, n ¼ 1) were detected.
DISCUSSION
Indication for CVAD Chemotherapy
Busch et al ▪ JVIR
477 (88.7%)
1,652 (95.4%)
Supportive therapy
27 (5.0%)
57 (3.3%)
Parenteral nutrition Access side
34 (6.3%)
23 (1.3%)
Left
414 (77.0%)
1,255 (72.5%)
Right
124 (23.0%)
477 (27.5%)
CVAD ¼ central venous access device; PUR ¼ polyurethane; SI ¼ silicone.
In this study, primary venographically guided implantation was technically successful in 97.15% of patients, comparable to the literature with an average of 98.9% (93.7%– 100%) according to a recent meta-analysis (2). Neither major complications nor procedure-related deaths occurred. Dwell times were significantly longer with SI catheters (174 d) compared with PUR catheters (159 d; P ¼ .003). This difference appears to arise from a change of paradigm.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
5
Table 2. Complications in Patients with Different TIVADs Complication
All Rate (n)
SI Catheter (Cook Vital-Port Mini Titanium) (n ¼ 538) Rate (n)
PUR Catheter (Titanium SlimPort) (n ¼ 1,732)
Catheter Days (Events/1,000 d)
Rate (n)
P
Catheter Days (Events/1,000 d)
Infections
4.67% (106)
4.64% (25)
0.15
4.68% (81)
0.19
1
Thrombotic dysfunction
4.27% (97)
3.53% (19)
0.11
4.50% (78)
0.19
.395
Thrombotic catheter occlusion
1.67% (38)
2.79% (15)
0.09
1.33% (23)
0.06
.035
Venous thrombosis
2.60% (59)
0.74% (4)
0.02
3.17% (55)
0.13
.003
3.30% (75)
6.69% (36)
0.22
2.25% (39)
0.09
< .001
0.84% (19) 2.47% (56)
3.35% (18) 3.35% (18)
0.11 0.11
0.06% (1) 2.19% (38)
< 0.01 0.09
< .001 .179
12.25% (278)
14.87% (80)
0.48
11.43% (198)
0.47
.040
Nonthrombotic dysfunction Material failures Other nonthrombotic dysfunction Total complication rate
Note–Complications classified according to the SIR guidelines (9): infections, including local infection (ie, erythema, induration, or purulence at the implantation site) and catheter-related sepsis (ie, positive blood culture and growth of organisms on the catheter tip, with no other source of infection); thrombotic dysfunction (ie, catheter-induced venous thrombosis or thrombotic catheter occlusion); or nonthrombotic dysfunction (ie, material failure, such as complete or partial catheter line fracture, or other dysfunction, such as catheter detachment, line migration, port hub rotation, wound dehiscence, or effects on the median nerve). PUR ¼ polyurethane; SI ¼ silicone; TIVAD ¼ totally implanted venous access device.
Table 3. Early-Onset and Late-Onset Complications in Patients with Different Port Devices Complication
All Rate
SI Catheter (Cook Vital-Port Mini Titanium)
PUR Catheter (Titanium SlimPort)
P
Rate
≤ 30 d, Rate (n)
> 30 d, Rate (n)
Rate
≤ 30 d, Rate (n)
> 30 d, Rate (n)
Infection
4.67%
4.64%
1.67% (9)
2.97% (16)
4.68%
1.27% (22)
3.41% (59)
Thrombosis Material failure
4.27% 0.84%
3.53% 3.35%
0.75% (4) 0.19% (1)
2.78% (15) 3.16% (17)
4.50% 0.06%
1.62% (28) 0.0% (0)
2.88% (50) 0.06% (1)
.395 < .001
2.47%
3.35%
0.57% (3)
2.19%
0.75% (13)
1.44% (25)
.179
12.25%
14.87%
3.64%
7.79%
.040
Others Total complication rate
3.16%
2.78% (15) 11.71%
11.43%
1
PUR ¼ polyurethane; SI ¼ silicone.
Figure 3. Venogram of the left arm and hemithorax of a 59year-old woman. Thrombosis (arrowheads) of the left basilic vein as well as the innominate vein is shown with multiple venous collaterals between the basilic vein, brachial vein, and axillary vein.
After an increased incidence of material failure observed in this department in 2011, it was strongly recommended to all subsequent patients to undergo device removal immediately after completion of therapy.
Figure 4. Fluoroscopic image in lateral projection of the left arm of a 70-year-old man. Partial catheter fracture is demonstrated by proximal leakage (arrow) of contrast agent.
Comparable to the current literature, the overall complication rate averaged 12.25% with a rate of 14.87% for SI catheters and 11.43% for PUR catheters (P ¼ .040) (2,6,10). The largest and most significant differences between SI and PUR devices were observed regarding material failure with
6 ▪ Complication Rates of Central Venous Access Catheters
Busch et al ▪ JVIR
Figure 5. Fluoroscopic images of the right arm (a) and chest (b) of a 56-year-old man with complete brachial catheter fracture (arrow, a) and asymptomatic catheter line migration to the left and right pulmonary arteries (arrowheads, b). The catheter fragment was successfully removed endovascularly.
a rate of 3.35% catheter fractures after a mean dwell time of 243 days with SI catheters but less than 0.1% after 1,109 days with PUR catheters (P < .001). In accordance with the assumption that TIVAD placement in the upper arm results in increased catheter angulation and stress at the vein entry, owing to continuous shearing at the brachial fascia, higher resistance flow, and mechanical tension, a uniform fracture pattern at the vein’s entry site was observed (3,11). In this respect, it has been noted that variations in catheter sizes (5 F for SI or 6 F for PUR) and longer average dwell times with SI catheters might bias these results, but other authors similarly observed a reduced mechanical stability of SI catheters compared with PUR catheters in single-lumen as well as multilumen catheters (12–14). As studies for this observation are limited, failure of SI hemodialysis catheters was attributed to weak spots after final processing owing to improper mixing of the SI matrix with insufficient removal of admixed air bubbles and unequal dispersion of barium sulfates (15). According to current research for equal stability, greater wall thickness is required with SI catheters compared with PUR catheters (16). The higher stiffness of PUR devices has advantages, such as allowing high-flow injections and lower risk of line displacement (10). However, there are also disadvantages, such as mechanical irritation of the vessel wall with increasing risk of thrombophlebitis and loss of function secondary to extravasal kinking (10,17). At the present time, despite many reports, there is neither evidence supporting superiority nor a specific recommendation regarding catheter materials, but with increasingly long dwell times, biodegradation may have an even larger impact on catheter maintenance and related complications (18,19). The most common complications were infections and thromboses. Despite the assumption that peripherally inserted TIVADs might be associated with higher thrombogenicity, regarding the greater intravenous catheter course as well as the proportionally higher intraluminal occlusion of smaller peripheral veins, an overall rate of 4.27% for catheter-related thrombosis was comparable to rates for chest ports (1.9%–4.8%) (3,20,21). Comparison of SI and PUR catheters revealed almost identical rates of infection (SI, 4.64%; PUR, 4.68%; P ¼ 1) and a slight, nonsignificant tendency of thrombotic dysfunction for PUR
catheters (SI, 3.53%; PUR, 4.50%; P ¼ .395). However, analyzing the different types of thrombotic incidences, PUR catheters were more prone to catheter-induced venous thromboses compared with thrombotic catheter occlusions (3.17% vs. 1.33%; P < .001), whereas the incidences were reversed in SI catheters (0.74% vs 2.79%; P < .001). In addition to the catheter materials, the different inner/outer diameters (SI, 1.0/1.7 mm [5 F]; PUR, 1.3/2.0 mm [6 F] may influence this observation. The almost equal susceptibility to infection of PUR and SI catheters is potentially linked to the similar rates of thrombotic incidence. However, other authors (14,22,23) reported on explicitly increasing catheter-related thrombosis and infections when using PUR catheters compared with SI catheters. This observation is based on the assumption that the in vivo use of PUR leads to significant biodegradation with more surface irregularities and subsequently increasing plasma protein absorption, which may induce thrombus formation. Moreover, there seems to be a close association of thrombogenicity and infectivity (24,25). The thrombus might serve as fertile soil for bacterial seeding, but the risk of developing thrombosis likewise increases substantially after an episode of infection. This study has some limitations. First, its retrospective design implies a lack of randomization and leads to a susceptibility to potential biases and confounders that may influence thrombogenicity and infectivity (eg, comorbidities, medications, coagulation status, type and consistency of catheter care regimens). Second, these data lack some information regarding accessed puncture site (ie, target vessel diameters, multiple puncture attempts during implantation, and history of thrombosis before TIVAD), but there is no evidence for association with catheter-related thrombosis (26). Third, only 1 brand of either SI or PUR catheter was considered; hence, these results cannot definitively be transferred to other manufacturers of SI or PUR catheters. Finally, in this department, TIVAD placement is usually performed under venography guidance; therefore, these results may not extend to ultrasound-guided vein access. In conclusion, the reported data suggest different risk profiles in SI catheters compared with PUR catheters. The 2 catheter types had similar infection rates; however, SI catheters were more prone to material failure and thrombotic
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
catheter occlusion, whereas PUR catheters had a higher susceptibility for catheter-induced venous thrombosis. Systematic investigations of the mechanical properties of explanted catheters are crucial to gain insight into the safety profile of catheter materials.
REFERENCES N, Castadot P, et al. Is radiologic placement of an arm 1. Marcy P-Y, Magne port mandatory in oncology patients? Analysis of a large bi-institutional experience. Cancer 2007; 110:2331–2338. 2. Burbridge B, Stoneham G, Lim HJ, Lee CH. Radiology-implanted arm ports: a review of the literature. JAVA 2015; 20:81–90. 3. Busch JD, Herrmann J, Heller F, et al. Follow-up of radiologically totally implanted central venous access ports of the upper arm: long-term complications in 127,750 catheter-days. AJR Am J Roentgenol 2012; 199:447–452. 4. Vardy J, Engelhardt K, Cox K, et al. Long-term outcome of radiologicalguided insertion of implanted central venous access port devices (CVAPD) for the delivery of chemotherapy in cancer patients: institutional experience and review of the literature. Br J Cancer 2004; 91: 1045–1049. 5. Zingg W, Cartier-Fassler V, Walder B. Central venous catheter-associated infections. Best Pract Res Clin Anaesthesiol 2008; 22:407–421. 6. Walser EM. Venous access ports: indications, implantation technique, follow-up, and complications. Cardiovasc Intervent Radiol 2012; 35: 751–764. 7. Baskin KM, Jimenez RM, Cahill AM, Jawad AF, Towbin RB. Cavoatrial junction and central venous anatomy: implications for central venous access tip position. J Vasc Interv Radiol 2008; 19:359–365. 8. Trautmann M, Jansen B, Frey P, Hummler H, Rasche M. Prevention of vascular catheter–associated infections. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2002; 45:907–924. [in German]. 9. Silberzweig JE, Sacks D, Khorsandi AS. Reporting standards for central venous access. J Vasc Interv Radiol 2003; 14:443–452. 10. Seckold T, Walker S, Dwyer T. A comparison of silicone and polyurethane PICC lines and postinsertion complication rates: a systematic review. J Vasc Access 2015; 16:167–177. 11. Balsorano P, Galducci G, De Fanti I, Evans SK, De Gaudio AR, Pelagatti C. Fractures of totally implantable central venous ports: more than fortuity. A three-year single center experience. J Vasc Access 2014; 15: 391–395.
7
12. Hwang FR, Stavropoulos SW, Shlansky-Goldberg RD, et al. Tunneled infusion catheter breakage: frequency and repair kit outcomes. J Vasc Interv Radiol 2008; 19:201–206. 13. Cohen AB, Dagli M, Stavropoulos SWJ, et al. Silicone and polyurethane tunneled infusion catheters: a comparison of durability and breakage rates. J Vasc Interv Radiol 2011; 22:638–641. 14. Wildgruber M, Lueg C, Borgmeyer S, et al. Polyurethane versus silicone catheters for central venous port devices implanted at the forearm. Eur J Cancer 2016; 59:113–124. 15. Weijmer MC, Kars SM, Wee ter PM. A scanning electron microscopy analysis of a spontaneous hemodialysis catheter fracture. Am J Kidney Dis 2001; 38:858–861. 16. Lenhart M, Sch€atzler S, Manke C, et al. Radiological placement of peripheral central venous access ports at the forearm. Technical results and long term outcome in 391 patients. Rofo 2010; 182:20–28. [in German]. 17. Pittiruti M, Brutti A, Celentano D, et al. Clinical experience with powerinjectable PICCs in intensive care patients. Crit Care 2012; 16:R21. 18. Busch JD, Vens M, Herrmann J, Adam G, Ittrich H. Material failure of silicone catheter lines: a retrospective review of partial and complete ruptures in 553 patients. AJR Am J Roentgenol 2016; 208:1–6. 19. Biffi R, Toro A, Pozzi S, Di Carlo I. Totally implantable vascular access devices 30 years after the first procedure. What has changed and what is still unsolved? Support Care Cancer 2014; 22:1705–1714. 20. Kock HJ, Pietsch M, Krause U, Wilke H, Eigler FW. Implantable vascular access systems: experience in 1500 patients with totally implanted central venous port systems. World J Surg 1998; 22:12–16. 21. Kuriakose P, Colon-Otero G, Paz-Fumagalli R. Risk of deep venous thrombosis associated with chest versus arm central venous subcutaneous port catheters: a 5-year single-institution retrospective study. J Vasc Interv Radiol 2002; 13:179–184. 22. Howard GT. Biodegradation of polyurethane: a review. Int Biodeterior Biodegradation 2002; 49:245–252. 23. Cauich-Rodríguez JV, Chan-Chan LH, Hernandez-Sanchez F, et al. Degradation of polyurethanes for cardiovascular applications. In: Pignatello R, editor. Advances in Biomaterials Science and Biomedical Applications. Rijeka, Croatia: InTech; 2013;51–82. 24. van Rooden CJ, Schippers EF. Infectious complications of central venous catheters increase the risk of catheter-related thrombosis in hematology patients: a prospective study. J Oncol 2005; 23:2655–2660. 25. Verbeke F, Haug U, Dhondt A, et al. The role of polymer surface degradation and barium sulphate release in the pathogenesis of catheterrelated infection. Nephrol Dial Transplant 2010; 25:1207–1213. 26. Goltz JP, Schmid JS, Ritter CO, et al. Identification of risk factors for catheter-related thrombosis in patients with totally implantable venous access ports in the forearm. J Vasc Access 2012; 13:79–85.
Complication Rates Observed in Silicone and Polyurethane Catheters of Totally Implanted Central Venous Access Devices Implanted in the Upper Arm Jasmin D. Busch, MD, Maren Vens, PhD, Catherine Mahler, MD, Jochen Herrmann, MD, Gerhard Adam, MD, and Harald Ittrich, MD
ABSTRACT Purpose: To present frequency and types of complications related to silicone (SI) versus polyurethane (PUR) catheters of totally implanted venous access devices (TIVADs) placed in the upper arm. Material and Methods: A cohort of 2,491 consecutive patients with TIVADs implanted between 2006 and 2015 was retrospectively analyzed. Complications were classified according to SIR guidelines. Pearson c2 test was used for categorical variables, and Student t test was used for continuous variables. Nominal P values were reported, and 2-sided P values < .05 were considered significant. Results: Of 2,270 patients meeting the inclusion criteria, 538 had an SI catheter, and 1,732 had a PUR catheter. Total dwell time was 584,853 catheter days. Mean total complication rate was 12.25% (SI, 14.87%; PUR, 11.43%; P ¼ .040). Subanalysis revealed significant differences for material failures (eg, catheter fracture [SI, 3.35%; PUR, 0.06%; P < .001] and thrombotic catheter occlusion/venous thromboses [SI, 2.79%/0.74%; PUR, 1.33%/3.17%; P < .001]) but nonsignificant differences for infections (eg, local infection and catheter-related sepsis [SI, 4.64%; PUR, 4.68%; P ¼ 1]) or other nonthrombotic dysfunctions (eg, catheter detachment, line migration, wound dehiscence [SI, 3.35%; PUR, 2.19%; P ¼ .179]). Conclusions: The reported data suggest different risk profiles in SI catheters compared with PUR catheters, with more material failures and thrombotic catheter occlusions in SI catheters and more venous thromboses in PUR catheters.
ABBREVIATIONS PUR ¼ polyurethane, SI ¼ silicone, TIVAD ¼ totally implanted venous access device
Reliable and convenient vascular access is an integral component of modern multimodality treatment, including chemotherapy, medication administration, parenteral nutrition, supportive treatment, and blood products. Totally implanted venous access devices (TIVADs) such as port systems have gained wide acceptance with increased
From the Section of Pediatric Radiology (J.D.B., J.H.), Department of Diagnostic and Interventional Radiology and Nuclear Medicine (G.A., H.I.), Department of Medical Biometry and Epidemiology (M.V.), and Department of Legal Medicine (C.M.), University Medical Center Hamburg-Eppendorf, Martinistrasse 52, Hamburg 20246, Germany. Received February 16, 2017; final revision received April 16, 2017; accepted April 24, 2017. Address correspondence to J.D.B.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–7 http://dx.doi.org/10.1016/j.jvir.2017.04.024
availability of catheters of different materials, mainly silicone (SI) or polyurethane (PUR) (1,2). Regardless of the benefit (3), in vivo TIVADs are exposed to multifactorial strain and are associated with complications, such as infection, thrombosis, or material failure (2,4–6). Although most complications are non–life-threatening events, they may lead to interruption of medical treatment. Essentially 2 types of TIVADs have been used since initiating percutaneous placement in our department in 2006 (Fig 1a, b). Despite a similar design of multicomponent devices used, there are distinct differences regarding the attached catheters. The Cook Vital-Port Mini Titanium (Cook, Inc, Bloomington, Indiana) is attached to an SI catheter, and the Titanium SlimPort (Bard Access Systems, Inc, Salt Lake City, Utah) is attached to a PUR catheter. The purpose of this study was to present the frequency and types of complications related to an SI catheter versus a PUR catheter for TIVAD placed in the upper arm over a 10-year period in a retrospective single-center observation.
2 ▪ Complication Rates of Central Venous Access Catheters
Busch et al ▪ JVIR
Figure 1. Differently designed TIVADs. (a) Cook Vital-Port Mini Titanium composed of a circular, tapered port hub (arrowhead) attached to a 5.0-F SI catheter and (b) Titanium SlimPort with a trapezoid port hub (arrowhead) attached to a 6.0-F polyurethane catheter.
MATERIALS AND METHODS Study Design The local institutional review board approved this retrospective single-center cohort study and waived the requirement for written informed consent. Between January 2006 and December 2015, 2,491 patients were referred to the interventional radiology department for fluoroscopic-guided implantation of a TIVAD in the upper arm. Data pertaining to the first device for each patient were analyzed. Further inclusion criteria were patients 18 years old at time of procedure, completely documented implantation of either a Cook Vital-Port Mini Titanium or a Titanium SlimPort, and documented follow-up time interval > 1 day. Patient data were censored at the time of first complication, device removal, date of last follow-up, or death. The end of the study was March 2016 (3 months after last TIVAD placement).
TIVAD Implantation Technique Hemostatic disorders (ie, platelet count < 50,000/μL and international normalized ratio > 1.5), bacteremia, and septicemia were regarded as absolute contraindications. Typically, the nondominant arm was used, with the exception of thrombosis of the implantation site, ipsilateral axillary lymph node dissection, or previously embedded pacemakers or automated implantable cardioverter-defibrillators. The basilic vein was the preferred and prevailing access site, but in individual cases of insurmountable puncture obstacles (eg, preexisting thrombosis), accessible veins were used alternatively. All procedures were performed by a radiologist with at least 2 years of practice in venous access procedures. Before implantation, deep vein thrombosis of the accessed side was evaluated for using venography (Integris Allura; Philips Healthcare, Best, Netherlands). Thereafter, under strictly aseptic conditions and after local anesthesia, contrast venography-guided vein access was achieved. Port hub placement was performed in the lower third of the upper arm, and the catheter line was tunneled from the puncture
site to the port pocket. In accordance with current recommendations, patients did not receive prophylactic antibiotics or routine anticoagulation during the follow-up period. After successful insertion with fluoroscopic confirmation of correct catheter tip positioning at the cavoatrial junction (defined as 2 vertebral bodies below the carina) (7), implanted systems were flushed with sterile heparinized saline (5 mL of a solution containing 100 IU/mL) (3,8).
Specification of TIVADs From January 2006 to February 2011, the Cook Vital-Port Mini Titanium was used in 553 consecutive patients. This device comprises a tapered titanium port hub (base radius 19.0 mm, height 7.2 mm) attached to a single-lumen 5.0-F SI catheter with inner/outer diameter of 1.0/1.7 mm. From January 2011 to December 2015, the Titanium SlimPort was used in 1,938 consecutive patients. This device comprises a tapered trapezoid titanium port hub (base 19.0 24 mm, height 9.4 mm) attached to a single-lumen 6.0-F PUR catheter with inner/outer diameter of 1.3/2.0 mm.
Follow-up Patients were assessed by an interventional radiologist 1–4 days after implantation and were immediately referred to the interventional radiology department if TIVAD-related complications were suspected. Medical data were obtained retrospectively by reviewing the radiologic information system (Centricity; GE Healthcare, Little Chalfont, United Kingdom). In patients who were lost to follow-up before removal of the TIVAD or who died within the study period, the last date of documented regular catheter usage was recorded.
Data Analysis The following characteristics were assessed: technical feasibility, successful implantation with correct catheter tip positioning and aspiration of blood as well as flushing of the
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
entire system with saline without any difficulty; dwell time, calculated from the day of placement until removal (in days) or, in patients who unable to be monitored, the last date of documented regular catheter usage; and complications, classified according to the Society of Interventional Radiology (SIR) guidelines (9). Events were additionally divided into early onset ( 30 d after implantation) or late onset (> 30 d after implantation) periprocedural complications.
Statistical Analysis Analyses were conducted with Microsoft Excel 2010 (Microsoft Corp, Redmond, Washington) and R (R Foundation for Statistical Computing, Vienna, Austria). For categorical variables, Pearson c2 test with Yates continuity correction was used. Student t test was used for continuous variables. If continuous variables were assumed to be nonnormal, the Wilcoxon rank sum test was used. Nominal P values are reported without correction for multiplicity. Two-sided P values < .05 were considered significant.
Figure 2. Flow chart of the study cohort.
3
RESULTS Technical Feasibility Technical feasibility is typically not related to the catheter material itself; hence, the entire cohort was considered for evaluation. Primary implantation was technically successful with a venographically guided approach in 97.15% (2,420 of 2,491) of patients. Initial insertion failed owing to a hematoma at the puncture site (n ¼ 20), venous spasm (n ¼ 16), central vein thrombosis (n ¼ 15), failed puncture in cases of lymph edema or obesity (n ¼ 11), and per patient request during the procedure (n ¼ 9). All procedures were completed without major complications.
Patient Cohort and Follow-up The inclusion criteria were met by 2,270 patients (Fig 2) with 538 SI catheters and 1,732 PUR catheters. The total dwell time accounted for 584,853 catheter days (SI, 166,824 d; PUR, 418,029 d). In 33.44% of patients (759 of 2,270; SI,
4 ▪ Complication Rates of Central Venous Access Catheters
228; PUR, 531), the TIVAD was removed after completion of treatment. Furthermore, 21.01% of patients (477 of 2,270; SI, 185; PUR, 292) died as a result of progression of their primary disease with a functional TIVAD still in place. No death occurred secondary to device-related complications. Detailed patient information is summarized in Table 1. Characteristics of patients receiving both devices were compared to evaluate for potential bias. There were no significant differences in age distribution (SI, 59.6 y; PUR, 58.1 y; P ¼ .056) and access sites (SI, 77.0%; PUR, 72.5%; P ¼ .055). Patients who received an SI catheter were significantly more frequently male (SI, 44.8%; PUR, 38.0%; P ¼ .006). Also, dwell times were significantly longer for SI catheters (174 d; interquartile range, 59–422 d; range, 2–2,515 d) compared with PUR catheters (159 d; interquartile range, 47–318 d; range, 2–1,728 d; P ¼ .003).
Complications The total mean complication rate was 12.25% (SI, 14.87%; PUR, 11.43%; P ¼ .040), including early-onset events ( 30 d) in 3.16% of patients with SI catheters and in 3.64% of patients with PUR catheters (P ¼ .025). Regarding
Table 1. Characteristics of Patients with a CVAD between 2006 and 2015 Characteristic
SI Catheter (Cook Vital-Port Mini Titanium)
PUR Catheter (Titanium SlimPort)
Number of patients
538
1,732
Male/female
241 (44.8%)/297 (55.2%)
659 (38.0%)/1,073 (62.0%)
Age, y, mean (range) Underlying disease
59.6 (19.5–88.5)
58.1 (18.2–91.7)
Breast cancer
117 (21.7%)
451 (26.0%)
Lung cancer
111 (20.6%)
187 (10.8%)
Head and neck cancer
108 (20.1%)
274 (15.8%)
Gastrointestinal cancer
74 (13.8%)
169 (9.8%)
Gynecologic cancer
31 (5.8%)
160 (9.2%)
Lymphoma/ leukemia
31 (5.8%)
147 (8.5%)
Urogenital cancer Other malignancies
10 (1.9%) 41 (7.6%)
60 (3.5%) 228 (13.2%)
Nonmalignant disease
15 (2.8%)
56 (3.2%)
diverse types of complications, SI and PUR catheters demonstrated nonsignificant differences for infection (P ¼ 1), thrombotic dysfunction (P ¼ .395), and other nonthrombotic dysfunction (P ¼ .179) but significant differences for material failure (P < .001). Detailed information about complications is listed in Tables 2 and 3.
Infections Overall, the mean infection rate was 4.67% (SI: 4.64%, PUR: 4.68%; P ¼ 1), including 1.37% early-onset infections (SI: 1.67%, PUR: 1.27%; P ¼ .6239).
Thrombotic Dysfunction In general, thrombotic dysfunction (Fig 3) occurred with a mean rate of 4.27% in all patients, with rates of 3.53% in patients with SI catheters and 4.50% in patients with PUR catheters (P ¼ .395). However, detailed consideration revealed significant differences with catheter-induced venous thromboses (SI, 0.74%; PUR, 3.17%; P < .001) and thrombotic catheter occlusions (SI, 2.79%; PUR, 1.33%; P < .001). SI catheters had more frequent thrombotic catheter occlusions (15 vs 4 of 538 patients), whereas PUR catheters were more prone to catheter-induced venous thrombosis (55 vs 23 of 1,732 patients). For both devices, there was no significant thrombotic predisposition regarding implantation site (SI, P ¼ .269; PUR, P ¼ .437).
Nonthrombotic Dysfunction The total rate of nonthrombotic dysfunction was 3.30%. In particular, regarding material failure, the differences between the 2 devices were greatest with 3.35% of complete (n ¼ 10) and partial fractures (n ¼ 8) of SI catheters after a mean dwell time of 243 days (Figs 4 and 5a, b). In contrast, only 1 patient with a PUR catheter presented with a partial fracture at time of removal after a dwell time of 1,109 days (0.06%; P < .001). Other nonthrombotic complications were noted for both catheters with an occurrence rate of 2.47%, with nonsignificant differences of 3.35% for SI devices and 2.19% for PUR devices (P ¼ .179). Complications such as port hub rotation (SI, n ¼ 8; PUR, n ¼ 6), line migration (SI, n ¼ 6; PUR, n ¼ 15), catheter detachment (SI, n ¼ 0; PUR, n ¼ 4), effects on the median nerve (SI, n ¼ 3; PUR, n ¼ 4), wound dehiscence (SI, n ¼ 1; PUR, n ¼ 8), and delayed allergic reaction (SI, n ¼ 0; PUR, n ¼ 1) were detected.
DISCUSSION
Indication for CVAD Chemotherapy
Busch et al ▪ JVIR
477 (88.7%)
1,652 (95.4%)
Supportive therapy
27 (5.0%)
57 (3.3%)
Parenteral nutrition Access side
34 (6.3%)
23 (1.3%)
Left
414 (77.0%)
1,255 (72.5%)
Right
124 (23.0%)
477 (27.5%)
CVAD ¼ central venous access device; PUR ¼ polyurethane; SI ¼ silicone.
In this study, primary venographically guided implantation was technically successful in 97.15% of patients, comparable to the literature with an average of 98.9% (93.7%– 100%) according to a recent meta-analysis (2). Neither major complications nor procedure-related deaths occurred. Dwell times were significantly longer with SI catheters (174 d) compared with PUR catheters (159 d; P ¼ .003). This difference appears to arise from a change of paradigm.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
5
Table 2. Complications in Patients with Different TIVADs Complication
All Rate (n)
SI Catheter (Cook Vital-Port Mini Titanium) (n ¼ 538) Rate (n)
PUR Catheter (Titanium SlimPort) (n ¼ 1,732)
Catheter Days (Events/1,000 d)
Rate (n)
P
Catheter Days (Events/1,000 d)
Infections
4.67% (106)
4.64% (25)
0.15
4.68% (81)
0.19
1
Thrombotic dysfunction
4.27% (97)
3.53% (19)
0.11
4.50% (78)
0.19
.395
Thrombotic catheter occlusion
1.67% (38)
2.79% (15)
0.09
1.33% (23)
0.06
.035
Venous thrombosis
2.60% (59)
0.74% (4)
0.02
3.17% (55)
0.13
.003
3.30% (75)
6.69% (36)
0.22
2.25% (39)
0.09
< .001
0.84% (19) 2.47% (56)
3.35% (18) 3.35% (18)
0.11 0.11
0.06% (1) 2.19% (38)
< 0.01 0.09
< .001 .179
12.25% (278)
14.87% (80)
0.48
11.43% (198)
0.47
.040
Nonthrombotic dysfunction Material failures Other nonthrombotic dysfunction Total complication rate
Note–Complications classified according to the SIR guidelines (9): infections, including local infection (ie, erythema, induration, or purulence at the implantation site) and catheter-related sepsis (ie, positive blood culture and growth of organisms on the catheter tip, with no other source of infection); thrombotic dysfunction (ie, catheter-induced venous thrombosis or thrombotic catheter occlusion); or nonthrombotic dysfunction (ie, material failure, such as complete or partial catheter line fracture, or other dysfunction, such as catheter detachment, line migration, port hub rotation, wound dehiscence, or effects on the median nerve). PUR ¼ polyurethane; SI ¼ silicone; TIVAD ¼ totally implanted venous access device.
Table 3. Early-Onset and Late-Onset Complications in Patients with Different Port Devices Complication
All Rate
SI Catheter (Cook Vital-Port Mini Titanium)
PUR Catheter (Titanium SlimPort)
P
Rate
≤ 30 d, Rate (n)
> 30 d, Rate (n)
Rate
≤ 30 d, Rate (n)
> 30 d, Rate (n)
Infection
4.67%
4.64%
1.67% (9)
2.97% (16)
4.68%
1.27% (22)
3.41% (59)
Thrombosis Material failure
4.27% 0.84%
3.53% 3.35%
0.75% (4) 0.19% (1)
2.78% (15) 3.16% (17)
4.50% 0.06%
1.62% (28) 0.0% (0)
2.88% (50) 0.06% (1)
.395 < .001
2.47%
3.35%
0.57% (3)
2.19%
0.75% (13)
1.44% (25)
.179
12.25%
14.87%
3.64%
7.79%
.040
Others Total complication rate
3.16%
2.78% (15) 11.71%
11.43%
1
PUR ¼ polyurethane; SI ¼ silicone.
Figure 3. Venogram of the left arm and hemithorax of a 59year-old woman. Thrombosis (arrowheads) of the left basilic vein as well as the innominate vein is shown with multiple venous collaterals between the basilic vein, brachial vein, and axillary vein.
After an increased incidence of material failure observed in this department in 2011, it was strongly recommended to all subsequent patients to undergo device removal immediately after completion of therapy.
Figure 4. Fluoroscopic image in lateral projection of the left arm of a 70-year-old man. Partial catheter fracture is demonstrated by proximal leakage (arrow) of contrast agent.
Comparable to the current literature, the overall complication rate averaged 12.25% with a rate of 14.87% for SI catheters and 11.43% for PUR catheters (P ¼ .040) (2,6,10). The largest and most significant differences between SI and PUR devices were observed regarding material failure with
6 ▪ Complication Rates of Central Venous Access Catheters
Busch et al ▪ JVIR
Figure 5. Fluoroscopic images of the right arm (a) and chest (b) of a 56-year-old man with complete brachial catheter fracture (arrow, a) and asymptomatic catheter line migration to the left and right pulmonary arteries (arrowheads, b). The catheter fragment was successfully removed endovascularly.
a rate of 3.35% catheter fractures after a mean dwell time of 243 days with SI catheters but less than 0.1% after 1,109 days with PUR catheters (P < .001). In accordance with the assumption that TIVAD placement in the upper arm results in increased catheter angulation and stress at the vein entry, owing to continuous shearing at the brachial fascia, higher resistance flow, and mechanical tension, a uniform fracture pattern at the vein’s entry site was observed (3,11). In this respect, it has been noted that variations in catheter sizes (5 F for SI or 6 F for PUR) and longer average dwell times with SI catheters might bias these results, but other authors similarly observed a reduced mechanical stability of SI catheters compared with PUR catheters in single-lumen as well as multilumen catheters (12–14). As studies for this observation are limited, failure of SI hemodialysis catheters was attributed to weak spots after final processing owing to improper mixing of the SI matrix with insufficient removal of admixed air bubbles and unequal dispersion of barium sulfates (15). According to current research for equal stability, greater wall thickness is required with SI catheters compared with PUR catheters (16). The higher stiffness of PUR devices has advantages, such as allowing high-flow injections and lower risk of line displacement (10). However, there are also disadvantages, such as mechanical irritation of the vessel wall with increasing risk of thrombophlebitis and loss of function secondary to extravasal kinking (10,17). At the present time, despite many reports, there is neither evidence supporting superiority nor a specific recommendation regarding catheter materials, but with increasingly long dwell times, biodegradation may have an even larger impact on catheter maintenance and related complications (18,19). The most common complications were infections and thromboses. Despite the assumption that peripherally inserted TIVADs might be associated with higher thrombogenicity, regarding the greater intravenous catheter course as well as the proportionally higher intraluminal occlusion of smaller peripheral veins, an overall rate of 4.27% for catheter-related thrombosis was comparable to rates for chest ports (1.9%–4.8%) (3,20,21). Comparison of SI and PUR catheters revealed almost identical rates of infection (SI, 4.64%; PUR, 4.68%; P ¼ 1) and a slight, nonsignificant tendency of thrombotic dysfunction for PUR
catheters (SI, 3.53%; PUR, 4.50%; P ¼ .395). However, analyzing the different types of thrombotic incidences, PUR catheters were more prone to catheter-induced venous thromboses compared with thrombotic catheter occlusions (3.17% vs. 1.33%; P < .001), whereas the incidences were reversed in SI catheters (0.74% vs 2.79%; P < .001). In addition to the catheter materials, the different inner/outer diameters (SI, 1.0/1.7 mm [5 F]; PUR, 1.3/2.0 mm [6 F] may influence this observation. The almost equal susceptibility to infection of PUR and SI catheters is potentially linked to the similar rates of thrombotic incidence. However, other authors (14,22,23) reported on explicitly increasing catheter-related thrombosis and infections when using PUR catheters compared with SI catheters. This observation is based on the assumption that the in vivo use of PUR leads to significant biodegradation with more surface irregularities and subsequently increasing plasma protein absorption, which may induce thrombus formation. Moreover, there seems to be a close association of thrombogenicity and infectivity (24,25). The thrombus might serve as fertile soil for bacterial seeding, but the risk of developing thrombosis likewise increases substantially after an episode of infection. This study has some limitations. First, its retrospective design implies a lack of randomization and leads to a susceptibility to potential biases and confounders that may influence thrombogenicity and infectivity (eg, comorbidities, medications, coagulation status, type and consistency of catheter care regimens). Second, these data lack some information regarding accessed puncture site (ie, target vessel diameters, multiple puncture attempts during implantation, and history of thrombosis before TIVAD), but there is no evidence for association with catheter-related thrombosis (26). Third, only 1 brand of either SI or PUR catheter was considered; hence, these results cannot definitively be transferred to other manufacturers of SI or PUR catheters. Finally, in this department, TIVAD placement is usually performed under venography guidance; therefore, these results may not extend to ultrasound-guided vein access. In conclusion, the reported data suggest different risk profiles in SI catheters compared with PUR catheters. The 2 catheter types had similar infection rates; however, SI catheters were more prone to material failure and thrombotic
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
catheter occlusion, whereas PUR catheters had a higher susceptibility for catheter-induced venous thrombosis. Systematic investigations of the mechanical properties of explanted catheters are crucial to gain insight into the safety profile of catheter materials.
REFERENCES N, Castadot P, et al. Is radiologic placement of an arm 1. Marcy P-Y, Magne port mandatory in oncology patients? Analysis of a large bi-institutional experience. Cancer 2007; 110:2331–2338. 2. Burbridge B, Stoneham G, Lim HJ, Lee CH. Radiology-implanted arm ports: a review of the literature. JAVA 2015; 20:81–90. 3. Busch JD, Herrmann J, Heller F, et al. Follow-up of radiologically totally implanted central venous access ports of the upper arm: long-term complications in 127,750 catheter-days. AJR Am J Roentgenol 2012; 199:447–452. 4. Vardy J, Engelhardt K, Cox K, et al. Long-term outcome of radiologicalguided insertion of implanted central venous access port devices (CVAPD) for the delivery of chemotherapy in cancer patients: institutional experience and review of the literature. Br J Cancer 2004; 91: 1045–1049. 5. Zingg W, Cartier-Fassler V, Walder B. Central venous catheter-associated infections. Best Pract Res Clin Anaesthesiol 2008; 22:407–421. 6. Walser EM. Venous access ports: indications, implantation technique, follow-up, and complications. Cardiovasc Intervent Radiol 2012; 35: 751–764. 7. Baskin KM, Jimenez RM, Cahill AM, Jawad AF, Towbin RB. Cavoatrial junction and central venous anatomy: implications for central venous access tip position. J Vasc Interv Radiol 2008; 19:359–365. 8. Trautmann M, Jansen B, Frey P, Hummler H, Rasche M. Prevention of vascular catheter–associated infections. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2002; 45:907–924. [in German]. 9. Silberzweig JE, Sacks D, Khorsandi AS. Reporting standards for central venous access. J Vasc Interv Radiol 2003; 14:443–452. 10. Seckold T, Walker S, Dwyer T. A comparison of silicone and polyurethane PICC lines and postinsertion complication rates: a systematic review. J Vasc Access 2015; 16:167–177. 11. Balsorano P, Galducci G, De Fanti I, Evans SK, De Gaudio AR, Pelagatti C. Fractures of totally implantable central venous ports: more than fortuity. A three-year single center experience. J Vasc Access 2014; 15: 391–395.
7
12. Hwang FR, Stavropoulos SW, Shlansky-Goldberg RD, et al. Tunneled infusion catheter breakage: frequency and repair kit outcomes. J Vasc Interv Radiol 2008; 19:201–206. 13. Cohen AB, Dagli M, Stavropoulos SWJ, et al. Silicone and polyurethane tunneled infusion catheters: a comparison of durability and breakage rates. J Vasc Interv Radiol 2011; 22:638–641. 14. Wildgruber M, Lueg C, Borgmeyer S, et al. Polyurethane versus silicone catheters for central venous port devices implanted at the forearm. Eur J Cancer 2016; 59:113–124. 15. Weijmer MC, Kars SM, Wee ter PM. A scanning electron microscopy analysis of a spontaneous hemodialysis catheter fracture. Am J Kidney Dis 2001; 38:858–861. 16. Lenhart M, Sch€atzler S, Manke C, et al. Radiological placement of peripheral central venous access ports at the forearm. Technical results and long term outcome in 391 patients. Rofo 2010; 182:20–28. [in German]. 17. Pittiruti M, Brutti A, Celentano D, et al. Clinical experience with powerinjectable PICCs in intensive care patients. Crit Care 2012; 16:R21. 18. Busch JD, Vens M, Herrmann J, Adam G, Ittrich H. Material failure of silicone catheter lines: a retrospective review of partial and complete ruptures in 553 patients. AJR Am J Roentgenol 2016; 208:1–6. 19. Biffi R, Toro A, Pozzi S, Di Carlo I. Totally implantable vascular access devices 30 years after the first procedure. What has changed and what is still unsolved? Support Care Cancer 2014; 22:1705–1714. 20. Kock HJ, Pietsch M, Krause U, Wilke H, Eigler FW. Implantable vascular access systems: experience in 1500 patients with totally implanted central venous port systems. World J Surg 1998; 22:12–16. 21. Kuriakose P, Colon-Otero G, Paz-Fumagalli R. Risk of deep venous thrombosis associated with chest versus arm central venous subcutaneous port catheters: a 5-year single-institution retrospective study. J Vasc Interv Radiol 2002; 13:179–184. 22. Howard GT. Biodegradation of polyurethane: a review. Int Biodeterior Biodegradation 2002; 49:245–252. 23. Cauich-Rodríguez JV, Chan-Chan LH, Hernandez-Sanchez F, et al. Degradation of polyurethanes for cardiovascular applications. In: Pignatello R, editor. Advances in Biomaterials Science and Biomedical Applications. Rijeka, Croatia: InTech; 2013;51–82. 24. van Rooden CJ, Schippers EF. Infectious complications of central venous catheters increase the risk of catheter-related thrombosis in hematology patients: a prospective study. J Oncol 2005; 23:2655–2660. 25. Verbeke F, Haug U, Dhondt A, et al. The role of polymer surface degradation and barium sulphate release in the pathogenesis of catheterrelated infection. Nephrol Dial Transplant 2010; 25:1207–1213. 26. Goltz JP, Schmid JS, Ritter CO, et al. Identification of risk factors for catheter-related thrombosis in patients with totally implantable venous access ports in the forearm. J Vasc Access 2012; 13:79–85.