Narrowed cephalic vein not usable for catheter insertion.
Di Carlo I, Cordio S, La Greca G, Privitera G, Russello D, Puleo S, Latteri F. Totally Implantable Venous Access Devices Implanted SurgicallyA Retrospective Study on Early and Late Complications. Arch Surg. 2001;136(9):1050-1053. doi:10.1001/archsurg.136.9.1050
The techniques used for the implantation of totally implantable venous access devices (TIVADs) are the percutaneous approach and surgical cutdown; however, the choice is still controversial.
The surgical cutdown approach may be beneficial to reduce the rate of complications.
A university hospital and a tertiary referral center.
Patients undergoing a TIVAD implant at the First Surgical Clinic of the University of Catania in Catania, Italy, between January 1995 and December 1999, were considered for the study. All of the devices were implanted in an operating room under fluoroscopic control. The vein of choice was the cephalic vein. When the cephalic vein was not suitable for implantation, the external jugular vein or the axillary vein and its branches were used. The percutaneous approach to the subclavian vein or internal jugular vein was considered a last resort to implant a catheter.
During the study period, 346 TIVADs were implanted in 344 patients. The procedure was performed with local anesthesia in 341 cases (98.5%), and only 2 patients (0.6%) required sedation for psychological reasons. Three patients (0.9%) had their TIVAD placed during a laparotomy. In 326 patients (94.2%), the devices were implanted in the cephalic vein. In the remaining cases, other veins were used with surgical cutdown. The mean time for the procedure was 15 minutes. Percutaneous access was never used, and no early mechanical complications were recorded. Only 6 patients (1.8%) in our study group had late complications (1 case of migration of the catheter, 2 cases of infection, and 3 cases of withdrawal occlusion). The catheter life ranged from 6 to 1487 days (mean time, 348 days).
Our results confirm the safety, speed, and low cost of the open cutdown technique. This surgical procedure avoids both early and late complications that frequently occur with percutaneous access. Surgical cutdown should be considered the technique of choice to implant the TIVAD, especially in cancer patients.
PERIPHERAL vascular access represents a significant problem in the administration of chemotherapy because of venous irritation and the need for multiple venipunctures. Totally implantable venous access devices (TIVADs) have become an important tool in the treatment of patients with malignant neoplasms.1 With the expanding use of the TIVAD, however, complications are being encountered more frequently. The percutaneous approach and surgical cutdown are the techniques used for TIVAD implantation. The relationship between both early and late mechanical complications and the technique of placement is reported in the literature.2,3 The purpose of this retrospective study was to analyze our experience with the TIVAD and to point out the factors influencing complications related to the technique of placement.
Between January 1995 and December 1999, at the First Surgical Clinic of the University of Catania in Catania, Italy, 346 TIVADs were implanted in 344 patients. The group consisted of 203 men and 141 women between the ages of 31 and 79 years. The indication for implantation was the infusion of intravenous chemotherapy in patients with solid tumors or hematologic disease. The only laboratory studies determined absolutely necessary were a complete blood cell count, including a platelet count, and coagulation tests. Bacteremia, septicemia, and disseminated intravascular coagulation were considered absolute contraindications for implantation.
In the present study, 62.1% of the TIVADs were of the same type (Port-a-Cath; SIMS Deltec Inc, St Paul, Minn). In the remaining 131 cases (37.9%), 2 different kinds of products (BardPort; Bard Access Systems, Salt Lake City, Utah, and Celsite; B/Braun Celsa, Chasseneuil, France) were used. The material of the injection port was titanium in all cases. In 246 cases (71.1%), the TIVAD catheters were made of silicone rubber; in 100 cases (28.9%), they were made of polyurethane.
All of the devices were implanted in the operating room under fluoroscopic control, even if the patient was treated and monitored in a day hospital setting. Those performing the procedure were surgeons or fellows in surgery. Perioperative antibiotic prophylaxis was performed as a single dose of cephalosporin given 10 minutes before the operation. All TIVADs were implanted using local anesthesia (usually 10 mL of 2% mepivacaine hydrochloride); sedation was reserved for patients for whom local anesthesia was not suitable. General anesthesia was used for the patients in whom TIVADs were placed during a major surgical procedure. The preferred method of implantation was to use the cephalic vein with a single skin incision at the Mohrenheim fossa. When the cephalic vein was absent or had been destroyed by previous chemotherapy and thus was not suitable for implantation, the external jugular vein or the axillary vein and its branches were used instead. The external jugular vein was used if clinically evident; otherwise the axillary vein and its branches were chosen. The percutaneous approach to the subclavian vein or internal jugular vein was considered a last resort to implant a catheter. Correct positioning of the distal tip of the catheter in the superior vena cava was carried out in all cases by fluoroscopic control. In particularly overweight patients, we eliminated half of the subcutaneous tissue to better localize the port for future punctures. Finally, a test puncture was performed to check patency and flow through the system. At the end of the procedure, a chest radiograph was obtained. Flushing with a solution of heparin sodium (5000 IU of heparin in 10 mL of isotonic saline) was performed every month.
From January 1995 to December 1999, 346 TIVADs were implanted in 344 patients. In all patients, the devices were inserted to infuse chemotherapy, in 98.9% for solid tumors and in 1.1% for hematologic disease (Table 1). Two hundred fifty-six TIVADs (74.1%) were implanted by a highly experienced surgeon, 40 (11.5%) were inserted by other surgeons, and 50 (14.4%) were implanted by fellows in surgery. The procedure was performed with local anesthesia in 341 cases (98.5%), and only 2 patients (0.6%) required sedation for psychological reasons. Three patients (0.9%) had their TIVAD placed in the vena cava and right gonadal vein during a laparotomy for colon cancer. In 326 patients (94.2%) the devices were implanted in the cephalic vein. In 15 patients (4.3%) the cephalic vein was not used: in 11 of these the vein was narrowed because of previous chemotherapy (Figure 1), in 2 patients the vein was hypoplastic, and in the remaining 2 the vein was absent. In these cases the catheter was surgically inserted using, respectively, the external jugular vein in 9 patients (2.6%), the axillary vein in 3 (0.9%), the coracobrachial vein in 2 (0.6%), and the internal jugular vein in 1 (0.3%). Percutaneous access was never used. The mean time for the procedure was 15 minutes (range, 9-45 minutes). A progressive increase in the number of implants was recorded from 1994 to 1999. In 36 patients (11.1%) we eliminated half of the subcutaneous tissue to better localize the port for its puncture. No early or late complications concerning the subcutaneous pocket were recorded in these subjects. Retrospective analysis of our study population showed that the catheter life ranged from 6 to 1487 days (mean time, 348 days). No early mechanical complications were recorded.
Only 6 patients (1.8%) in our study had late complications. Six days after implantation, 1 patient's TIVAD would not infuse; a chest x-ray showed a migration of the catheter into the subcutaneous pocket. The catheter was replaced, and the patient has used the TIVAD since then. Two patients (0.52%) experienced infection of the subcutaneous pocket, and after curative antibiotic therapy, the system was removed and a new TIVAD was implanted. The causative agent was coagulase-negative Staphylococcus aureus in both patients. Finally, in 3 patients (0.87%) withdrawal occlusion was recorded; x-rays never localized malposition of the tip of the catheter or a fibrin sleeve. Because these patients required only infusion, the systems were left in place. The TIVAD was treated with a heparin solution flush without specific anticoagulant treatment. One patient died after 210 days with the system in place; 2 patients have continued to use the TIVAD 378 and 679 days after implantation, respectively. The catheters of 2 of these TIVADs are made of silicone, and 1 of polyurethane.
Vascular access for patients requiring long-term drug delivery was a serious and challenging problem until the development of TIVADs. Since the first implant, performed by Niederhuber and colleagues4 in 1982, these systems have increasingly been used in the field of oncology. They achieve safe and less painful vascular access, facilitate treatment of many medical disorders, and improve patients' quality of life by giving them unrestricted mobility and freedom in their activities. These devices can be implanted through either a surgical or percutaneous procedure.1- 3 Although many surgical teams use percutaneous access for placement of the TIVAD, both early and late complications are frequently observed (Table 2). Pneumothorax is the most frequent complication and has the greatest clinical, economic, and psychological implications. The best time to implant the TIVAD is during the principal oncologic surgical procedure, but this decision must often be delayed until after the pathologist and oncologist consult. In this case a second surgical step is required, with consequential psychological effects on the patient. Therefore, the surgeon who performs the TIVAD implantation should use a technique with a low rate of complications to avoid further psychological trauma. In our experience no early mechanical complications occurred; we particularly want to stress the absence of pneumothorax, a clinical condition related to the percutaneous approach. Patients who undergo the percutaneous approach for TIVAD implantation and develop pneumothorax increase the costs of their procedure by US $535 if they are managed with observation; when tube thoracostomy is required, the cost for each patient becomes much higher (US $1827).5 Pneumothorax occurs during the subclavian vein puncture, and the percutaneous approach with the internal jugular vein has been suggested by some authors to avoid this complication.6,7 However, arterial puncture and related complications have been reported with this method. In our opinion, the physician performing the implant should be fully versed in the alternative puncture technique, but the impossibility of using the superficial venous system should not be considered justification for the percutaneous approach, as suggested by others.1 With our patients, it was always possible to use the surgical technique with a vein and avoid the risk of complications related to the percutaneous approach.
Some authors use the learning curve to explain the incidence of pneumothorax with the percutaneous method.2,5 Cancer patients should be treated with particular care on the basis of their often negative clinical course. For this reason, surgery fellows should be taught the surgical method, and the percutaneous approach should be reserved for other diseases.
Last but not least, the surgical approach prevents "pinch-off" and its relatively disastrous complications due to the migration of the fractured catheter. On the basis of a radiographic scale, Klotz et al8 suggest the removal of the catheter system within 6 months when confronted with the typical sign of a pinch-off. In our opinion, when the pinch-off sign is present, the catheter should be removed immediately to avoid fatal complications.9
Migration of the catheter tip during insertion is a consequence of not using fluoroscopy during the surgical implant. The use of fluoroscopy during catheter placement seems important not only to avoid immediate migration of the catheter but also to prevent malfunction. In fact, malfunction (defined as resistance to flush or infuse a substance, complete inability to infuse a substance, or resistance to blood aspiration) may occur whenever the catheter is in place and is related to the anomalous position of the catheter tip.10 Placement of TIVADs in an operating room or outpatient clinic setting without a means to localize the tip of the catheter should be avoided.11 In our 3 patients with malfunction, the tip of the catheter was correctly positioned. It appears that the withdrawal occlusion was due to long-lasting total parenteral nutrition infusion without adequate postinfusion treatment.
There are 2 types of catheter-associated thrombosis: small thrombosis limited to the tip of the catheter, characterized by an inability to withdraw blood from the system with unimpaired infusion into the system, and large thrombosis of the catheterized veins.12 The incidence of this complication is not reported in the literature because it usually is not clinically relevant. In our experience, no clinical evidence of thrombosis was recorded, and only 1 case of tip occlusion was found. The clinical course of this patient was particularly benign and did not include therapy. This catheter tip thrombosis is currently considered the first step toward fibrin sleeve formation along the catheter, causing withdrawal occlusion. Experimental results suggest that the sleeve around a central venous catheter is not a fibrin sleeve but a stable cellular-collagen tissue covered by the endothelium, mainly formed by smooth muscle cells migrating from the injured vein wall into the early pericatheter thrombus.13 For this reason, anticoagulants such as warfarin sodium need further investigation before standardization for clinical use.1
Port pocket infection, caused by gram-positive cocci, suggests either direct inoculation or migration of the organism along the access needle as the primary mechanism. Two of our patients developed this complication during a long period in which the system was implemented in a different department and flushing with a heparin solution was performed by another team of medical nurses. With a standardized technique and correct care and maintenance, the rate of infectious complications could be zero.
Our results confirm the safety, speed, and low cost of the open cutdown technique. This surgical procedure avoids both early and late complications that frequently occur with percutaneous access. The surgical cutdown should be considered the preferred technique to implant the TIVAD, especially in cancer patients. The pathological composition of the fibrin sleeve needs further investigation before standardization of anticoagulants for clinical use. Further reduction of complication rates is feasible through the combined efforts of medical oncologists, surgeons, and nursing staff.
Corresponding author: Isidoro Di Carlo, MD, via Regina Margherita n 180, 98034 Francavilla (ME), Italy (e-mail: firstname.lastname@example.org).
Reprints: Isidoro Di Carlo, MD, First Surgical Clinic, University of Catania, 95100 Catania, Italy.