Outpatient Percutaneous Central Venous Access in Cancer Patients J. Ralph Broadwater, MD, Michael A. Henderson, MD, John L. Bell, MD, Michael J. Edwards, MD, G. Jeffrey Smith, MD, David R. McCready, MO, Richard S. Swanson, MD, Mark E. R. Hardy, MD, Robert R. Shenk, MD, Millie Lawson, RN, David M. Ota, MD, Charles M. Balch, MD, Houston,Texas
A 1-year experience of percutaneous subclavian catheterization in outpatients with cancer was reviewed to document reliability, safety, and cost. There were 763 catheter insertions attempted with prospective documentation of complications in 664 consecutive patients. Catheter insertion was successful in 722 attempts ( 9 5 % ) . There were only 13 pneumothoraces ( 2 % ) . Thirty catheters required repositioning ( 4 % ) . The average catheter duration was 191 days (range: 0 to 892 days). Fifty-six catheters (8%) were removed because of suspected infection. Documented catheter sepsis occurred in 21 patients (3%) ; catheter site infection occurred in 8 patients ( 1 % ) . Thus, only 0.22 infections per catheter year occurred during this 382 catheter-year experience. The estimated cost of catheter insertion was $562, which is one-third the estimated cost for tunneled catheters ($1,403) and for reservoir devices ( $ 1 , 7 3 8 ) . In our experience, percutaneous subclavian catheterization is a reliable, cost-effective method compared with tunneled or reservoir devices, with an equivalent incidence of catheter-related infections. The cornerstone of our success with this program is a staff dedicated to catheter care and intensive patient education. In centers where a large number of patients require central venous access, pereutaneous catheterization should be the technique of choice.
From the Department of General Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. Requests for reprints should be addressed to J. Ralph Broadwater, MD, Department of Surgery, University of Arkansas for Medical Sciences, 4301 West Markham, Slot #725, Little Rock, Arkansas 72205. Presented at the 42rid Annual Meeting of the Southwestern Surgical Congress, La Quinta, California, April 22-25, 1990.
676
THE AMERICAN JOURNAL OF SURGERY
entral venous catheterization has become a valuable technique facilitating the safe delivery of chemoC therapeutic agents in the management of cancer patients. It also provides a route for antibiotic therapy, intravenous hyperalimentation, and blood withdrawal for laboratory testing. Traditionally, tunneled catheters (e.g., Hickman, Broviac, and Raaf catheters) have been used because they are thought to reduce the incidence of catheter sepsis [1-3]. More recently, subcutaneous reservoirs (e.g., Infusaport and Portacath, among others) have been used to avoid external catheters and allow for a more active lifestyle [4-6]. In 1976, the University of Texas M. D. Anderson Cancer Center instituted the use of percutaneous subclavian catheters for outpatient care of cancer patients. The program was developed because of the manpower requirements for intraoperative insertion of tunneled catheters, and the impression that patients frequently had complications with tunneled catheters that required removal during therapy. A noncuffed silicone elastomer catheter (Centrasil catheter, Baxter Travenol, Deerfield, IL) was used. Initial analysis suggested that this method was safe, with a low rate of infection or other catheter-related complications [7]. The purpose of this study was to analyze the reliability, safety, and cost of percutaneous subclavian catheterization in outpatients with cancer. PATIENTS AND METHODS Insertion of percutaneous subclavian catheters was attempted in 664 consecutive outpatients at the University of Texas M. D. Anderson Cancer Center in 1985; these cases form the basis of this report. All patients were treated for malignant disease, and most catheters were inserted to administer chemotherapy. Catheter use was not restricted, and the catheters were used for chemotherapy, blood transfusions, antibiotic therapy, blood chemistry analysis, and intravenous hyperalimentation. Since 1984, data concerning insertion complications and catheter removal have been recorded in the patients' medical records prospectively in the infusion therapy clinic. Diagnostic and demographic information and information on catheter exchanges, septic episodes, and reasons for catheter removal were obtained from a retrospective review of the medical record. For each patient, information on sex, diagnosis, and associated medical conditions (diabetes, steroid use, collagen vascular disease, presence of tracheostomy) was obtained. The following data on insertion were collected: date of insertion, side of insertion, patient factors complicating insertion (previous neck surgery, tracheostomy, cervical disease, and mediastinal disease), insertion complications (pneumothorax, medi-
VOLUME 160
DECEMBER 1990
PERCUTANEOUS SUBCLAVIAN CATHETERIZATION IN CANCER PATIENTS
astinal hematoma, lymphocele, and catheter malposition), and management of complications. Data gathered on catheter removal included indication for removal, resuits of catheter tip culture, exit site culture, blood culture, number of catheter exchanges, and date of removal. Cost was estimated by obtaining the charges for an average percutaneous subclavian catheter insertion. A similar cost estimate was made for tunneled and reservoir devices. Catheter insertion: All catheters are inserted by surgical house staff. A modified Seldinger technique is utilized for infraclavicular insertion of the silicone catheter into the subclavian vein. Sterile conditions are maintained, but masks and gowns are not used. A chest radiograph (posteroanterior and lateral views) is obtained after the catheter is inserted to confirm that it is positioned properly and that no complications have occurred. The patient and family are required to return to the infusion therapy clinic the next day for repeated instruction on catheter care. Another clinic visit is required to demonstrate that they can care for the catheter. Insertion complications are managed as follows. Catheter malposition is corrected with fluoroscopic guidance by the radiologist. Patients who have evidence of mediastinal hematoma or pneumothorax are admitted to the hospital and serial chest radiographs are obtained. A pneumothorax that is symptomatic or increasing in size requires a tube thoracostomy. Patients in whom catheter insertion fails are examined and their charts reviewed. If there is suspicion of subclavian vein occlusion, a radionuclide venogram is obtained to document patency of the superior vena cava. Patient education: The infusion therapy team at our institution is responsible for the care of all intravenous catheters and maintains an outpatient clinic staffed by six full-time registered nurses and two technicians. Patients and their families are interviewed by the infusion therapy nurses to ensure that they are capable of caring for catheters at home. Using audiovisual aids, the nurses instruct the patients and families in sterile techniques, heparin flushing, dressing changes, recognition of problems, obtaining help, and acquiring supplies. Patients and families are required to demonstrate proficiency in catheter care and dressing changes. Platelet count, coagulation profile, chest radiographs, and informed consent are also obtained prior to catheter insertion. Catheter management and wound care: Patients return monthly for catheter and wound inspection. Catheters require daily injection of 1 to 2 mL of heparin solution (100 U/mL) and a weekly dressing change. Patients are instructed to return to the clinic if there is any evidence of catheter malfunction, difficulty injecting heparin, evidence of wound infection at the catheter exit site, or fever. A malfunctioning catheter is managed by exchange over a guide wire. Catheter occlusion or thrombosis is managed by injection of 0.5 to 1 mL of urokinase solution (5,000 U/mL), or catheter removal and insertion of a new catheter. Patients who have fever that is believed to be catheter-related are admitted to the hospital and treated with intravenous antibiotics. Catheters are not
TABLE I Catheter Insertion Complications
Complication
Insertion Attempts* No. %
Inability to insert Pneumothorax Mediastinal hematoma Catheter malposition
41 13 8 30
5 2 1 4
Total
92
12
*Of the 763 catheter insertion attempts, 722 (95 %) were successful.
TABLE lI Indications for Catheter Removal Indication for Removal
No.
%
Treatment completed Death Sepsis suspected Sepsis proven Catheter thrombosis Catheter removed accidentally Site infection Malposition Malfunction Insertion complication Reason unknown
310 140 56 21 21 11 8 7 7 4 34
43 19 8 3 3 2 1 1 1 1 5
Total
619
87"
"At the time of chart review, 103 (14%) catheters remained in place.
routinely removed unless there is evidence of exit site infection or unless infection is associated with hypotension or is refractory to intravenous antibiotics. Cultures from the catheter tip, skin exit site, and blood are obtained if there is suspicion of sepsis or site infection. Catheter-related sepsis is defined as any positive blood culture with no other identified source of infection. Statistical analysis: Data were analyzed using d-Base III Plus. Statistical analysis was performed using SPSSPC Plus. Catheter survival was calculated by the KaplanMeier method. RESULTS A total of 763 catheter insertions were attempted in 664 consecutive outpatients. The mean age of all patients was 48 years. Five hundred eighty-one patients (88%) required only 1 catheter. Seventy-three (12%) had 2 catheters, while 10 patients (2%) had 3 or more catheters during their treatment. Insertion was successful in 722 of the 763 attempts (95%). Insertion complications are summarized in Table I. Pneumothorax occurred in 13 patients (2%). Eleven patients who had a pneumothorax required hospital admission. Of these, only six (0.1%) required tube thoracostomies; five patients were observed with serial chest radiographs. Two patients with small, stable pneumoth-
T H E A M E R I C A N J O U R N A L OF S U R G E R Y
V O L U M E 160
D E C E M B E R 1990
677
BROADWATER ET AL
Estimated insertion charges for a silicone percutaneous catheter are $562. This compares with estimated insertion charges of $1,403 for a tunneled catheter and $1,738 for a reservoir device.
100
80
60
40
20
I
.
.
.
.
.
0
.
.
.
.
i
.
.
.
.
.
.
.
.
.
300
i
.
.
.
.
.
.
.
.
600
.
i
900
DAYS Figure 1. Percutaneous subclavian catheter survival.
TABLE I l l
Estimated Cost by Central Venous Catheterization Technique"
Charge (in dollars) Insertion charges Physician Anesthesia Operating room
Percutaneous Silicone Catheter
Tunneled Catheter
Reservoir
Supplies
265 NA NA 77 37 112 23 48
450 350 336 181 NA NA 54 32
500 350 336 181 NA NA 339 32
Total
562
1,403
t, 738
Radiology Nursing Clinic visit +
Catheter
NA = not applicable. *Cost estimated from routine hospital charges. ~lnsertion requires four clinic visits.
oraces were managed as outpatients. Mediastinal hematoma occurred in eight patients as documented by radiographic examination; none required surgical intervention. Thirty catheters (4%) required repositioning after initial insertion. There were no lymphoceles. Reasons for catheter removal are summarized in Table II. Only 21 patients had documented infection (3%); 56 (8%) catheters were removed for suspected sepsis. Eight patients (1%) had infection at the exit site. Mean catheter duration in all patients was 193 days (range: 0 to 892 days) (Figure 1). No catheter-related death occurred. Analysis of cost indicated that the percutaneous technique is the least expensive method of central venous access in cancer patients. Catheter insertion costs were estimated for three different methods of central venous access at our institution and are summarized in Table III. 678
THE AMERICAN JOURNAL OF SURGERY
COMMENTS Cancer patients requiring long-term central venous access can be treated safely as outpatients using percutaneous subclavian catheters. Catheter insertion was successful in 95% of patients, and pneumothorax occurred in only 2%. Thirty catheters (4%) were placed in an incorrect position and required fluoroscopic repositioning. Our results compare favorably with the literature, which describes an incidence of pneumothorax ranging from 0% to 12.4%, and of catheter malposition in 4.5% to 32% of patients. Dudrick and co-workers' [8] landmark contribution in 1968 showed the utility of parenteral nutrition and established the need for central venous access. Initially, catheters were placed percutaneously in the subclavian vein for total parenteral nutrition (TPN). The indications for central venous access were expanded as the insertion technique was standardized. Ryan et al [9] reviewed insertion and infectious complications prospectively in 200 patients requiring central venous access for TPN nutrition. Catheter-related sepsis occurred in 7% of catheters, but in only 3% of catheters where strict aseptic technique was used. The Centers for Disease Control surveyed 12 major medical centers using central venous access for TPN and found a sepsis rate of 7.3% in 1,363 patients [10]. In 1969, the longest reported catheter use was 90 days, with an average of 24 days [11]. Standard teaching was that patients could have central venous access as inpatients if strict aseptic technique was used to manage the catheter. Broviac et al [1] first described the use of a silicone rubber catheter for prolonged parenteral alimentation in 1973. This catheter contained a Dacron cuff and was placed through a subcutaneous tunnel. Hickman et al [2], in 1979, described a modification to a larger bore catheter. Since these two reports, a number of experiences with tunneled catheters have been reported [13,12]. A summary of the literature reveals a mean catheter duration of approximately 124 days. Our percutaneously inserted non-cuffed catheters had a mean catheter lifespan of 193 days. Tunneled catheters have been reported to reduce the incidence of catheter-related sepsis. Press et al [12] reviewed 992 patients with 1,088 catheters, and reported 143 catheter-related infections and 4 catheter-related deaths. Catheter-relat~ septicemia occurred in 4% of patients and exit site or tunnel infections in 6%. Catheterrelated sepsis occurred in only 3% of patients with percutaneous catheters in our series, and only 1% of patients had exit-site infections. There were no catheter-related deaths. The infection rate of tunneled catheters in Press' series was 4 per 100 catheter days, which is greater than the 0.06 infections per 100 catheter days in our study. In our review of 382 catheter-years' experience, there were only 0.22 infections per catheter year. Tunneled catheters
VOLUME 160
DECEMBER 1990
PERCUTANEOUSSUBCLAVIANCATHETERIZATIONIN CANCERPATIENTS
offer no distinct advantage over percutaneous catheters in reducing the incidence of sepsis and are more costly to insert. Totally implanted central venous devices were first described by Niederhuber et al [4] in 1982. These reservoirs have no external catheter, require no scheduled dressing changes, and need only monthly injection of heparin. The initial description by Niederhuber and coworkers of 20 patients with central venous access reservoirs revealed a single incidence of pneumothorax and an average catheter lifespan of 274 days. They reported no infectious complications. However, as experience with reservoir devices has accumulated, reports of infectious complications have increased, and range from 2% to 11% [5,6,13,14].
Our data indicate that percutaneous insertion of silicone subclavian catheters is an effective method for longterm central venous access in cancer patients. Insertion complications were uncommon, catheter lifespan was similar to reported rates for tunneled or reservoir catheters, and the incidence of catheter-related infectious complications was no different from that for tunneled or reservoir devices. It is important to emphasize that these long catheter durations and low infection rates occurred in a high-risk group of patients with malignancies requiring intensive chemotherapy. A specific outpatient program for the insertion and care of percutaneous subclavian central venous catheters was developed at the M. D. Anderson Cancer Center because the large number of patients who need access for chemotherapy administration would require an excessive amount of operating room and physician time for catheter insertion. In addition, central venous catheters may require removal for catheter and infectious complications; removal of tunneled and reservoir catheters requires an operative procedure. At first, long-term central venous access was obtained with silicone elastomer central venous catheters inserted via the brachial vein [15,16]. Median catheter duration time was 238 days, and although catheter-related sepsis rates were low, the brachial catheters had a significant incidence of peripheral thrombophlebitis (6%). For this reason, the silicone catheters were inserted percutaneously in the subclavian vein for long-term central venous access. Two other series have reported long-term use of percutaneous catheters. In 1985, Slater et al [17] reported a consecutive experience in 140 cancer patients. Average catheter duration was 190 days, and the incidence of infectious complications was 13%. Periman et al [18] described their experience with 133 catheters. Mean catheter duration was 112 days, and infectious complications occurred in 8%. Keohane et al [19] conducted a controlled trial between tunneled and percutaneous catheters. These investigators showed no significant difference in the catheter sepsis rate between the two methods when a nurse dedicated to catheter care helped manage the patients' catheters. Analysis of our data and a comparison of the results obtained with tunneled catheters and reservoir devices
reveal no distinct advantage of tunneled cuffed catheters over percutaneous silicone catheters. Percutaneous catheters have the advantage of outpatient insertion. Removal is simple if a catheter-related complication occurs. However, they have the disadvantage of requiring daily heparin injections and weekly dressing changes. Tunneled catheters have the disadvantage of requiring a more invasive procedure for removal if a catheter-related complication occurs. No external dressing changes are required, but daily heparin injection is recommended. Reservoir devices require operating room insertion, and infectious complications require operative removal. They have the advantage of no external catheter, no dressing changes, and require only once-monthly heparin injection. Reservoir devices may be useful in children and patients with more active lifestyles, particularly patients who enjoy swimming or other water sports. Percutaneous catheters are the most cost-effective method of central venous insertion at our institution. Insertion cost was only one-third of that for tunneled or reservoir devices. Our institution has a daily clinic with six full-time nurses because of the large number of patients requiring central venous access. An important element in the success of outpatient percutaneous catheterization is a detailed educational program involving the patients and their families. In centers where such a program is achievable, percutaneous subclavian catheters should be the technique of choice because they cost less, can be inserted without operating room time, and are easily removed if complications occur.
REFERENCES 1. Broviac JW, Cole J J, Scribner BH. A silicon rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 1973; 136: 602-6. 2. Hickrnan RO, Buckner CD, Clift RD, et al. A modified right atrial catheter for access to the venoussystem in marrow transplant recipients. Surg Gynecol Obstet 1979; 148: 871-5. 3. Wagman LD, Kirkemo A, Johnston MR. Venous access: a prospective, randomized study of the Hickman catheter. Surgery 1984; 95: 303-8. 4. Niederhuber JE, Ensminger WE, Gyves JW, et al. Totally implanted venous and arterial system to replace external catheters in cancer treatment. Surgery 1982; 92: 706-12. 5. Bothe A, Piccione W, Ambrosino J J, et al. Implantable central venous access system. Am J Surg 1984; 147: 565-9. 6. Lokich J J, Bothe A, Moore C. Complicationsand management of implanted venous access catheters. J Clin Oncol 1985; 3:710-7. 7. McCredie KB, Lawson M. Percutaneous insertion of silicone central venouscatheters for long-term intravenous access in cancer patients. IM 1984; 5: 100-5. 8. Dudrick SJ, Wilmore DW, Vars HM, et al. Long term parenteral nutrition with growth developmentand positivenitrogen balance. Surgery 1968; 64: 134-42. 9. Ryan JA, Abel RM, Abbot WM, et al. Catheter complications in total parenteral nutrition. A prospectivestudy of 200 consecutive patients. N Engl J Med 1974; 290: 757-61. 10. Goldman DA, Maki DG. Deliverof the patient. In: Symposium on Total Nutrition, AMA 1972, p. 132-140. 11. Wilmore DW, Dudrick SJ. Safe long term venous catheterization. Arch Surg 1969; 98: 256-8. 12. Press OW, Ramsey PG, Larson EB, et al. Hickman catheter infections in patients with malignancies. Medicine (Baltimore)
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990 679
BROADWATER ET AL
1984; 63: 189-200. 13. Lokich J J, Becker B. Subclavian vein thrombosis in patients treated with infusion chemotherapy for advanced malignancies. Cancer 1983; 52: 1586-9. 14. Stanislav GV, Fitzgibbons RJ, Bailey RT, et al. Reliability of implantable central venous access devices in patients with cancer. Arch Surg 1987; 122: 1280-4. 15. Legha SS, Haq M, Rabinowits M. Evaluation of silicone elastomer catheters for long-term intravenous chemotherapy. Arch Intern Med 1985; 145: 1208-11. 16. Bottino J, McCredie KB, Groschel DHM, et al. Long-term intravenous therapy with peripherally inserted silicone elastomer central venous catheters in patients with malignant diseases. Cancer 1979; 43: 1937-43. 17. Slater H, Goldfarb W, Jacob HE, et al. Experience with longterm outpatient venous access utilizing percutaneously placed silicone elastomer catheters. Cancer 1985; 56: 2074-7. 18. Periman P, Lane R, Cherni-Smith R. Vascular access for cancer chemotherapy. Tex Med 1986; 82: 44-5. 19. Keohane PP, Attril H, Northover J. Effect of catheter tunneling and a nutrition nurse on catheter sepsisduring parenteral nutrition. A controlled trial. Lancet 1983; 2: 1388-90.
DISCUSSION Gilehrist L. Jackson (Houston, TX): The outpatient technique is obviously cost effective. Done under local anesthesia, it does require, as the author pointed out, a very skilled nursing team. We have seen all different types of options for venous access, from arterial venous fistulas with polytetrafluorethylene conduits and various types of named catheters and ports. We have a new Groshong catheter with a valve that doesn't require heparin flush. Perhaps there is less infection with tunneled catheters, and the ports have the advantage of only monthly injection, but there may be a higher incidence of wound infection. When we first used the Centrasil catheter at our hospital, there were some technical problems with the catheter, including breakage. I think the Anderson experience shows that these technical problems have been resolved.
680
Carey P. Page (San Antonio, TX): Another potential advantage of the single-lumen percutaneous catheter was not emphasized in the presentation: in patients with suspected catheter-related infection who are not overtly septic, the existing catheter may be exchanged over a guidewire, both renewing access and confirming or eliminating the suspected diagnosis.
Steve Men-ell (Salt Lake City, UT): My question relates to another alternative for central access, which is the Intrasil catheter. This type of catheter may be introduced through a peripheral arm vein, then floated centrally into the superior vena cava. We have managed more than 600 Intrasil catheters over the past 4 or 5 years, with results similar to those you presented, but sparing all the morbidity related to central punctures. Eighty-six percent of the Intrasil catheters in our series were used for the entire length of time they were needed without complications. What are your views regarding the use of this type of catheter? Also, what proportion of the central venous catheters in your series were used for the entire time period central access was desired without complications that required replacement of the catheter? J. Ralph Broadwater (closing): Dr. Jackson, we currently recommend the use of percutaneous catheters in patients who require less than 6 months of therapy. We would recommend reservoir devices in patients who desire more active lifestyles, such as in people who want to swim or in children. I think it is important to study the length of time and the frequency of heparinization. The peripheral catheter was used initially in Houston and its success was the reason that the central catheters were used percutaneously. One of the reasons this was changed was because a fairly high incidence of phlebitis was found with the brachial catheter. In this report, approximately 20% of the patients required more than a single catheter.
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990
A 1-year experience of percutaneous subclavian catheterization in outpatients with cancer was reviewed to document reliability, safety, and cost. There were 763 catheter insertions attempted with prospective documentation of complications in 664 consecutive patients. Catheter insertion was successful in 722 attempts ( 9 5 % ) . There were only 13 pneumothoraces ( 2 % ) . Thirty catheters required repositioning ( 4 % ) . The average catheter duration was 191 days (range: 0 to 892 days). Fifty-six catheters (8%) were removed because of suspected infection. Documented catheter sepsis occurred in 21 patients (3%) ; catheter site infection occurred in 8 patients ( 1 % ) . Thus, only 0.22 infections per catheter year occurred during this 382 catheter-year experience. The estimated cost of catheter insertion was $562, which is one-third the estimated cost for tunneled catheters ($1,403) and for reservoir devices ( $ 1 , 7 3 8 ) . In our experience, percutaneous subclavian catheterization is a reliable, cost-effective method compared with tunneled or reservoir devices, with an equivalent incidence of catheter-related infections. The cornerstone of our success with this program is a staff dedicated to catheter care and intensive patient education. In centers where a large number of patients require central venous access, pereutaneous catheterization should be the technique of choice.
From the Department of General Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. Requests for reprints should be addressed to J. Ralph Broadwater, MD, Department of Surgery, University of Arkansas for Medical Sciences, 4301 West Markham, Slot #725, Little Rock, Arkansas 72205. Presented at the 42rid Annual Meeting of the Southwestern Surgical Congress, La Quinta, California, April 22-25, 1990.
676
THE AMERICAN JOURNAL OF SURGERY
entral venous catheterization has become a valuable technique facilitating the safe delivery of chemoC therapeutic agents in the management of cancer patients. It also provides a route for antibiotic therapy, intravenous hyperalimentation, and blood withdrawal for laboratory testing. Traditionally, tunneled catheters (e.g., Hickman, Broviac, and Raaf catheters) have been used because they are thought to reduce the incidence of catheter sepsis [1-3]. More recently, subcutaneous reservoirs (e.g., Infusaport and Portacath, among others) have been used to avoid external catheters and allow for a more active lifestyle [4-6]. In 1976, the University of Texas M. D. Anderson Cancer Center instituted the use of percutaneous subclavian catheters for outpatient care of cancer patients. The program was developed because of the manpower requirements for intraoperative insertion of tunneled catheters, and the impression that patients frequently had complications with tunneled catheters that required removal during therapy. A noncuffed silicone elastomer catheter (Centrasil catheter, Baxter Travenol, Deerfield, IL) was used. Initial analysis suggested that this method was safe, with a low rate of infection or other catheter-related complications [7]. The purpose of this study was to analyze the reliability, safety, and cost of percutaneous subclavian catheterization in outpatients with cancer. PATIENTS AND METHODS Insertion of percutaneous subclavian catheters was attempted in 664 consecutive outpatients at the University of Texas M. D. Anderson Cancer Center in 1985; these cases form the basis of this report. All patients were treated for malignant disease, and most catheters were inserted to administer chemotherapy. Catheter use was not restricted, and the catheters were used for chemotherapy, blood transfusions, antibiotic therapy, blood chemistry analysis, and intravenous hyperalimentation. Since 1984, data concerning insertion complications and catheter removal have been recorded in the patients' medical records prospectively in the infusion therapy clinic. Diagnostic and demographic information and information on catheter exchanges, septic episodes, and reasons for catheter removal were obtained from a retrospective review of the medical record. For each patient, information on sex, diagnosis, and associated medical conditions (diabetes, steroid use, collagen vascular disease, presence of tracheostomy) was obtained. The following data on insertion were collected: date of insertion, side of insertion, patient factors complicating insertion (previous neck surgery, tracheostomy, cervical disease, and mediastinal disease), insertion complications (pneumothorax, medi-
VOLUME 160
DECEMBER 1990
PERCUTANEOUS SUBCLAVIAN CATHETERIZATION IN CANCER PATIENTS
astinal hematoma, lymphocele, and catheter malposition), and management of complications. Data gathered on catheter removal included indication for removal, resuits of catheter tip culture, exit site culture, blood culture, number of catheter exchanges, and date of removal. Cost was estimated by obtaining the charges for an average percutaneous subclavian catheter insertion. A similar cost estimate was made for tunneled and reservoir devices. Catheter insertion: All catheters are inserted by surgical house staff. A modified Seldinger technique is utilized for infraclavicular insertion of the silicone catheter into the subclavian vein. Sterile conditions are maintained, but masks and gowns are not used. A chest radiograph (posteroanterior and lateral views) is obtained after the catheter is inserted to confirm that it is positioned properly and that no complications have occurred. The patient and family are required to return to the infusion therapy clinic the next day for repeated instruction on catheter care. Another clinic visit is required to demonstrate that they can care for the catheter. Insertion complications are managed as follows. Catheter malposition is corrected with fluoroscopic guidance by the radiologist. Patients who have evidence of mediastinal hematoma or pneumothorax are admitted to the hospital and serial chest radiographs are obtained. A pneumothorax that is symptomatic or increasing in size requires a tube thoracostomy. Patients in whom catheter insertion fails are examined and their charts reviewed. If there is suspicion of subclavian vein occlusion, a radionuclide venogram is obtained to document patency of the superior vena cava. Patient education: The infusion therapy team at our institution is responsible for the care of all intravenous catheters and maintains an outpatient clinic staffed by six full-time registered nurses and two technicians. Patients and their families are interviewed by the infusion therapy nurses to ensure that they are capable of caring for catheters at home. Using audiovisual aids, the nurses instruct the patients and families in sterile techniques, heparin flushing, dressing changes, recognition of problems, obtaining help, and acquiring supplies. Patients and families are required to demonstrate proficiency in catheter care and dressing changes. Platelet count, coagulation profile, chest radiographs, and informed consent are also obtained prior to catheter insertion. Catheter management and wound care: Patients return monthly for catheter and wound inspection. Catheters require daily injection of 1 to 2 mL of heparin solution (100 U/mL) and a weekly dressing change. Patients are instructed to return to the clinic if there is any evidence of catheter malfunction, difficulty injecting heparin, evidence of wound infection at the catheter exit site, or fever. A malfunctioning catheter is managed by exchange over a guide wire. Catheter occlusion or thrombosis is managed by injection of 0.5 to 1 mL of urokinase solution (5,000 U/mL), or catheter removal and insertion of a new catheter. Patients who have fever that is believed to be catheter-related are admitted to the hospital and treated with intravenous antibiotics. Catheters are not
TABLE I Catheter Insertion Complications
Complication
Insertion Attempts* No. %
Inability to insert Pneumothorax Mediastinal hematoma Catheter malposition
41 13 8 30
5 2 1 4
Total
92
12
*Of the 763 catheter insertion attempts, 722 (95 %) were successful.
TABLE lI Indications for Catheter Removal Indication for Removal
No.
%
Treatment completed Death Sepsis suspected Sepsis proven Catheter thrombosis Catheter removed accidentally Site infection Malposition Malfunction Insertion complication Reason unknown
310 140 56 21 21 11 8 7 7 4 34
43 19 8 3 3 2 1 1 1 1 5
Total
619
87"
"At the time of chart review, 103 (14%) catheters remained in place.
routinely removed unless there is evidence of exit site infection or unless infection is associated with hypotension or is refractory to intravenous antibiotics. Cultures from the catheter tip, skin exit site, and blood are obtained if there is suspicion of sepsis or site infection. Catheter-related sepsis is defined as any positive blood culture with no other identified source of infection. Statistical analysis: Data were analyzed using d-Base III Plus. Statistical analysis was performed using SPSSPC Plus. Catheter survival was calculated by the KaplanMeier method. RESULTS A total of 763 catheter insertions were attempted in 664 consecutive outpatients. The mean age of all patients was 48 years. Five hundred eighty-one patients (88%) required only 1 catheter. Seventy-three (12%) had 2 catheters, while 10 patients (2%) had 3 or more catheters during their treatment. Insertion was successful in 722 of the 763 attempts (95%). Insertion complications are summarized in Table I. Pneumothorax occurred in 13 patients (2%). Eleven patients who had a pneumothorax required hospital admission. Of these, only six (0.1%) required tube thoracostomies; five patients were observed with serial chest radiographs. Two patients with small, stable pneumoth-
T H E A M E R I C A N J O U R N A L OF S U R G E R Y
V O L U M E 160
D E C E M B E R 1990
677
BROADWATER ET AL
Estimated insertion charges for a silicone percutaneous catheter are $562. This compares with estimated insertion charges of $1,403 for a tunneled catheter and $1,738 for a reservoir device.
100
80
60
40
20
I
.
.
.
.
.
0
.
.
.
.
i
.
.
.
.
.
.
.
.
.
300
i
.
.
.
.
.
.
.
.
600
.
i
900
DAYS Figure 1. Percutaneous subclavian catheter survival.
TABLE I l l
Estimated Cost by Central Venous Catheterization Technique"
Charge (in dollars) Insertion charges Physician Anesthesia Operating room
Percutaneous Silicone Catheter
Tunneled Catheter
Reservoir
Supplies
265 NA NA 77 37 112 23 48
450 350 336 181 NA NA 54 32
500 350 336 181 NA NA 339 32
Total
562
1,403
t, 738
Radiology Nursing Clinic visit +
Catheter
NA = not applicable. *Cost estimated from routine hospital charges. ~lnsertion requires four clinic visits.
oraces were managed as outpatients. Mediastinal hematoma occurred in eight patients as documented by radiographic examination; none required surgical intervention. Thirty catheters (4%) required repositioning after initial insertion. There were no lymphoceles. Reasons for catheter removal are summarized in Table II. Only 21 patients had documented infection (3%); 56 (8%) catheters were removed for suspected sepsis. Eight patients (1%) had infection at the exit site. Mean catheter duration in all patients was 193 days (range: 0 to 892 days) (Figure 1). No catheter-related death occurred. Analysis of cost indicated that the percutaneous technique is the least expensive method of central venous access in cancer patients. Catheter insertion costs were estimated for three different methods of central venous access at our institution and are summarized in Table III. 678
THE AMERICAN JOURNAL OF SURGERY
COMMENTS Cancer patients requiring long-term central venous access can be treated safely as outpatients using percutaneous subclavian catheters. Catheter insertion was successful in 95% of patients, and pneumothorax occurred in only 2%. Thirty catheters (4%) were placed in an incorrect position and required fluoroscopic repositioning. Our results compare favorably with the literature, which describes an incidence of pneumothorax ranging from 0% to 12.4%, and of catheter malposition in 4.5% to 32% of patients. Dudrick and co-workers' [8] landmark contribution in 1968 showed the utility of parenteral nutrition and established the need for central venous access. Initially, catheters were placed percutaneously in the subclavian vein for total parenteral nutrition (TPN). The indications for central venous access were expanded as the insertion technique was standardized. Ryan et al [9] reviewed insertion and infectious complications prospectively in 200 patients requiring central venous access for TPN nutrition. Catheter-related sepsis occurred in 7% of catheters, but in only 3% of catheters where strict aseptic technique was used. The Centers for Disease Control surveyed 12 major medical centers using central venous access for TPN and found a sepsis rate of 7.3% in 1,363 patients [10]. In 1969, the longest reported catheter use was 90 days, with an average of 24 days [11]. Standard teaching was that patients could have central venous access as inpatients if strict aseptic technique was used to manage the catheter. Broviac et al [1] first described the use of a silicone rubber catheter for prolonged parenteral alimentation in 1973. This catheter contained a Dacron cuff and was placed through a subcutaneous tunnel. Hickman et al [2], in 1979, described a modification to a larger bore catheter. Since these two reports, a number of experiences with tunneled catheters have been reported [13,12]. A summary of the literature reveals a mean catheter duration of approximately 124 days. Our percutaneously inserted non-cuffed catheters had a mean catheter lifespan of 193 days. Tunneled catheters have been reported to reduce the incidence of catheter-related sepsis. Press et al [12] reviewed 992 patients with 1,088 catheters, and reported 143 catheter-related infections and 4 catheter-related deaths. Catheter-relat~ septicemia occurred in 4% of patients and exit site or tunnel infections in 6%. Catheterrelated sepsis occurred in only 3% of patients with percutaneous catheters in our series, and only 1% of patients had exit-site infections. There were no catheter-related deaths. The infection rate of tunneled catheters in Press' series was 4 per 100 catheter days, which is greater than the 0.06 infections per 100 catheter days in our study. In our review of 382 catheter-years' experience, there were only 0.22 infections per catheter year. Tunneled catheters
VOLUME 160
DECEMBER 1990
PERCUTANEOUSSUBCLAVIANCATHETERIZATIONIN CANCERPATIENTS
offer no distinct advantage over percutaneous catheters in reducing the incidence of sepsis and are more costly to insert. Totally implanted central venous devices were first described by Niederhuber et al [4] in 1982. These reservoirs have no external catheter, require no scheduled dressing changes, and need only monthly injection of heparin. The initial description by Niederhuber and coworkers of 20 patients with central venous access reservoirs revealed a single incidence of pneumothorax and an average catheter lifespan of 274 days. They reported no infectious complications. However, as experience with reservoir devices has accumulated, reports of infectious complications have increased, and range from 2% to 11% [5,6,13,14].
Our data indicate that percutaneous insertion of silicone subclavian catheters is an effective method for longterm central venous access in cancer patients. Insertion complications were uncommon, catheter lifespan was similar to reported rates for tunneled or reservoir catheters, and the incidence of catheter-related infectious complications was no different from that for tunneled or reservoir devices. It is important to emphasize that these long catheter durations and low infection rates occurred in a high-risk group of patients with malignancies requiring intensive chemotherapy. A specific outpatient program for the insertion and care of percutaneous subclavian central venous catheters was developed at the M. D. Anderson Cancer Center because the large number of patients who need access for chemotherapy administration would require an excessive amount of operating room and physician time for catheter insertion. In addition, central venous catheters may require removal for catheter and infectious complications; removal of tunneled and reservoir catheters requires an operative procedure. At first, long-term central venous access was obtained with silicone elastomer central venous catheters inserted via the brachial vein [15,16]. Median catheter duration time was 238 days, and although catheter-related sepsis rates were low, the brachial catheters had a significant incidence of peripheral thrombophlebitis (6%). For this reason, the silicone catheters were inserted percutaneously in the subclavian vein for long-term central venous access. Two other series have reported long-term use of percutaneous catheters. In 1985, Slater et al [17] reported a consecutive experience in 140 cancer patients. Average catheter duration was 190 days, and the incidence of infectious complications was 13%. Periman et al [18] described their experience with 133 catheters. Mean catheter duration was 112 days, and infectious complications occurred in 8%. Keohane et al [19] conducted a controlled trial between tunneled and percutaneous catheters. These investigators showed no significant difference in the catheter sepsis rate between the two methods when a nurse dedicated to catheter care helped manage the patients' catheters. Analysis of our data and a comparison of the results obtained with tunneled catheters and reservoir devices
reveal no distinct advantage of tunneled cuffed catheters over percutaneous silicone catheters. Percutaneous catheters have the advantage of outpatient insertion. Removal is simple if a catheter-related complication occurs. However, they have the disadvantage of requiring daily heparin injections and weekly dressing changes. Tunneled catheters have the disadvantage of requiring a more invasive procedure for removal if a catheter-related complication occurs. No external dressing changes are required, but daily heparin injection is recommended. Reservoir devices require operating room insertion, and infectious complications require operative removal. They have the advantage of no external catheter, no dressing changes, and require only once-monthly heparin injection. Reservoir devices may be useful in children and patients with more active lifestyles, particularly patients who enjoy swimming or other water sports. Percutaneous catheters are the most cost-effective method of central venous insertion at our institution. Insertion cost was only one-third of that for tunneled or reservoir devices. Our institution has a daily clinic with six full-time nurses because of the large number of patients requiring central venous access. An important element in the success of outpatient percutaneous catheterization is a detailed educational program involving the patients and their families. In centers where such a program is achievable, percutaneous subclavian catheters should be the technique of choice because they cost less, can be inserted without operating room time, and are easily removed if complications occur.
REFERENCES 1. Broviac JW, Cole J J, Scribner BH. A silicon rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 1973; 136: 602-6. 2. Hickrnan RO, Buckner CD, Clift RD, et al. A modified right atrial catheter for access to the venoussystem in marrow transplant recipients. Surg Gynecol Obstet 1979; 148: 871-5. 3. Wagman LD, Kirkemo A, Johnston MR. Venous access: a prospective, randomized study of the Hickman catheter. Surgery 1984; 95: 303-8. 4. Niederhuber JE, Ensminger WE, Gyves JW, et al. Totally implanted venous and arterial system to replace external catheters in cancer treatment. Surgery 1982; 92: 706-12. 5. Bothe A, Piccione W, Ambrosino J J, et al. Implantable central venous access system. Am J Surg 1984; 147: 565-9. 6. Lokich J J, Bothe A, Moore C. Complicationsand management of implanted venous access catheters. J Clin Oncol 1985; 3:710-7. 7. McCredie KB, Lawson M. Percutaneous insertion of silicone central venouscatheters for long-term intravenous access in cancer patients. IM 1984; 5: 100-5. 8. Dudrick SJ, Wilmore DW, Vars HM, et al. Long term parenteral nutrition with growth developmentand positivenitrogen balance. Surgery 1968; 64: 134-42. 9. Ryan JA, Abel RM, Abbot WM, et al. Catheter complications in total parenteral nutrition. A prospectivestudy of 200 consecutive patients. N Engl J Med 1974; 290: 757-61. 10. Goldman DA, Maki DG. Deliverof the patient. In: Symposium on Total Nutrition, AMA 1972, p. 132-140. 11. Wilmore DW, Dudrick SJ. Safe long term venous catheterization. Arch Surg 1969; 98: 256-8. 12. Press OW, Ramsey PG, Larson EB, et al. Hickman catheter infections in patients with malignancies. Medicine (Baltimore)
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990 679
BROADWATER ET AL
1984; 63: 189-200. 13. Lokich J J, Becker B. Subclavian vein thrombosis in patients treated with infusion chemotherapy for advanced malignancies. Cancer 1983; 52: 1586-9. 14. Stanislav GV, Fitzgibbons RJ, Bailey RT, et al. Reliability of implantable central venous access devices in patients with cancer. Arch Surg 1987; 122: 1280-4. 15. Legha SS, Haq M, Rabinowits M. Evaluation of silicone elastomer catheters for long-term intravenous chemotherapy. Arch Intern Med 1985; 145: 1208-11. 16. Bottino J, McCredie KB, Groschel DHM, et al. Long-term intravenous therapy with peripherally inserted silicone elastomer central venous catheters in patients with malignant diseases. Cancer 1979; 43: 1937-43. 17. Slater H, Goldfarb W, Jacob HE, et al. Experience with longterm outpatient venous access utilizing percutaneously placed silicone elastomer catheters. Cancer 1985; 56: 2074-7. 18. Periman P, Lane R, Cherni-Smith R. Vascular access for cancer chemotherapy. Tex Med 1986; 82: 44-5. 19. Keohane PP, Attril H, Northover J. Effect of catheter tunneling and a nutrition nurse on catheter sepsisduring parenteral nutrition. A controlled trial. Lancet 1983; 2: 1388-90.
DISCUSSION Gilehrist L. Jackson (Houston, TX): The outpatient technique is obviously cost effective. Done under local anesthesia, it does require, as the author pointed out, a very skilled nursing team. We have seen all different types of options for venous access, from arterial venous fistulas with polytetrafluorethylene conduits and various types of named catheters and ports. We have a new Groshong catheter with a valve that doesn't require heparin flush. Perhaps there is less infection with tunneled catheters, and the ports have the advantage of only monthly injection, but there may be a higher incidence of wound infection. When we first used the Centrasil catheter at our hospital, there were some technical problems with the catheter, including breakage. I think the Anderson experience shows that these technical problems have been resolved.
680
Carey P. Page (San Antonio, TX): Another potential advantage of the single-lumen percutaneous catheter was not emphasized in the presentation: in patients with suspected catheter-related infection who are not overtly septic, the existing catheter may be exchanged over a guidewire, both renewing access and confirming or eliminating the suspected diagnosis.
Steve Men-ell (Salt Lake City, UT): My question relates to another alternative for central access, which is the Intrasil catheter. This type of catheter may be introduced through a peripheral arm vein, then floated centrally into the superior vena cava. We have managed more than 600 Intrasil catheters over the past 4 or 5 years, with results similar to those you presented, but sparing all the morbidity related to central punctures. Eighty-six percent of the Intrasil catheters in our series were used for the entire length of time they were needed without complications. What are your views regarding the use of this type of catheter? Also, what proportion of the central venous catheters in your series were used for the entire time period central access was desired without complications that required replacement of the catheter? J. Ralph Broadwater (closing): Dr. Jackson, we currently recommend the use of percutaneous catheters in patients who require less than 6 months of therapy. We would recommend reservoir devices in patients who desire more active lifestyles, such as in people who want to swim or in children. I think it is important to study the length of time and the frequency of heparinization. The peripheral catheter was used initially in Houston and its success was the reason that the central catheters were used percutaneously. One of the reasons this was changed was because a fairly high incidence of phlebitis was found with the brachial catheter. In this report, approximately 20% of the patients required more than a single catheter.
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990
