J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION

Transcription

1 VOLUME 22 NUMBER 13 JULY JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T Prophylactic Urokinase in the Management of Long- Term Venous Access Devices in Children: A Children s Oncology Group Study Peter W. Dillon, Gary R. Jones, Holly A. Bagnall-Reeb, Jonathon D. Buckley, Eugene S. Wiener, and Gerald M. Haase From the Department of Surgery, Division of Pediatric Surgery, Penn State College of Medicine, Hershey; Department of Surgery, Children s Hospital of Pittsburgh, Pittsburgh, PA; Department of Pediatrics, Division of Pediatric Hematology/Oncology, Oregon Health Sciences University, Portland, OR; Department of Preventive Medicine, University of Southern California Keck School of Medicine, Los Angeles, CA; Department of Pediatric Surgery, The Children s Hospital, Denver, CO. Submitted July 7, 2003; accepted April 10, This study was funded in part by a grant from Abbott Laboratories to the Children s Cancer Group. G.R.J., G.M.H., and H.A.B.-R. served as consultants to Abbott Laboratories. No individuals received direct remuneration from Abbott Laboratories during the conduct of this study. Authors disclosures of potential conflicts of interest are found at the end of this article. Address reprint requests to Peter Dillon, MD, Division of Pediatric Surgery, MC H113, 500 University Dr, Penn State Milton S. Hershey Medical Center, Hershey, PA 17033; pdillon1@psu.edu CC: smason@ childrensoncologygroup.org by American Society of Clinical Oncology X/04/ /$20.00 DOI: /JCO A B S T R A C T Purpose Infection and thrombosis are serious complications of long-term vascular access devices in children undergoing chemotherapy. Since routine fibrinolytic therapy may decrease these complications, the purpose of this study was to compare the efficacy of an every-2-week administration of urokinase with standard heparin flushes in reducing the incidence of device-related infections and occlusions. Materials and Methods This study was a prospective, randomized phase III multicenter trial conducted by the Children s Cancer Group, in which patients with implantable ports or tunneled catheters received either urokinase or heparin every 2 weeks for 12 months. Study end points were time to first occlusion or time to first device-related infection. Results Five hundred seventy-seven patients from 29 institutions were enrolled, of whom 51% had external catheters and 49% had ports. Urokinase administration resulted in fewer occlusive events than heparin (23% v 31%; P.02), a longer time to first occlusive event (log-rank analysis, P.006), and a 1.6-fold difference in the rate of occlusive events (Poisson regression, P.003). Similar results were noted when comparing ports and tunneled catheters. The urokinase group also had a 1.4-fold difference in the rate of infection (Poisson regression, P.05) and longer time to first infection (log-rank, P.07), but the difference was significant only in tunneled catheters. Conclusion Urokinase administration every 2 weeks significantly affects the rate of occlusive events in ports and tunneled catheters and of infectious events in external catheters compared with heparin administration. J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION Indwelling central venous access devices have become the cornerstone of supportive care and therapy in the treatment of most childhood malignancies. They have solved the challenge of establishing long-term venous access, have altered dramatically the nature of chemotherapy administration, and have improved significantly the quality of life of children undergoing treatment. Despite their widespread acceptance, device infection and thrombosis are the two most common and serious complications associated with their use. It has been reported that 20% to 35% of tunneled catheters are removed before completion of therapy due to infection and thrombotic occlusion with risk of substantial patient morbidity. 1,2 Studies have shown that not only thrombotic occlusion but also catheterrelated infection may be due to fibrin deposition associated with catheters. 3,4 Clinical use of fibrinolytic agents such as urokinase has been shown to be highly efficacious in reopening catheters occluded by thrombus. 5,6 Several small studies have suggested a role for use of periodic fibrinolytic therapy in decreasing catheter-related complications including infection However, 2718

2 Urokinase Prophylaxis in Pediatric Vascular Devices there are no published reports of studies that involved multiple institutions in which this strategy was used in a large population. The purpose of this randomized Children s Cancer Group (CGC) phase III multicenter trial was to compare the efficacy of an every-2-week (14 day) instillation of urokinase versus heparin for reducing the incidence of catheter-related infections and occlusions in central venous access devices in children with malignancies. PATIENTS AND METHODS This study was designed as a prospective randomized phase III multicenter trial consisting of CGC patients who had central venous access devices and were undergoing active chemotherapy treatments. Patients were stratified by type of central venous device implantable intravenous access device (port) or an externally tunneled catheter (Broviac/Hickman C.R. Bard Inc, Salt Lake City, UT) to receive either urokinase or heparin instillations. Patient enrollment and data collection occurred during an 18- month period from July 1997 to December 1998 at participating CCG institutions. Patients were assigned randomly to have each lumen of their central venous catheter flushed every 2 weeks with either urokinase 5,000 units/ml (Abbokinase Open-Cath, 9,000 units/1.8 ml; Abbott Laboratories, Abbott Park, IL) or heparin sodium (Heplock USP; Leo Pharmaceutical Products, Ballerup, Denmark) 100 units/ml in a volume sufficient to fill the entire device. Study drug was required to remain within the device lumen(s) for a minimum of 1 hour to a maximum of 14 days (2 weeks). The variation in dwell time was determined by patient convenience and clinical status. Because catheter care protocols differed between institutions, individual institutional practices for catheter care and heparin flushing were permitted between study drug instillations. The initial dose of study drug was administered no later than 7 days after device insertion. All study treatments were scheduled every 14 days but were allowed to occur within an 8 to 20 day timeframe between administrations. Inclusion criteria included patients 21 years of age or younger who had a central venous access device inserted no more than 7 days before enrollment on the study. Patients had to have an anticipated life expectancy for the duration of the study, were expected to require use of the device for a minimum of 6 months, and had to be able to return to the primary institution for evaluation and study drug administration. Patients or their legally authorized representatives had to sign the informed consent form in conformance with the Declaration of Helsinki. Withdrawal from the study was permitted at any time. Institutional review board approval was obtained at each treating institution. Exclusion criteria included patients who had an occluded venous access device at the time of study enrollment, patients with any existing local or systemic infection, patients with a Groshong device (C.R. Bard Inc), or patients receiving anticoagulation drugs other than routine heparin flush or agents for the prophylaxis of catheter-related infection or thrombosis. The use of bactrim/septra/oral antifungal agents was allowed. Additional exclusion criteria also included body weight less than 5 kg, age younger than 3 months, and previous enrollment on the study. Devices were inserted according to individual institutional protocols and practices. Patients were then followed up every 2 weeks for assessment and study drug administration by a nurse, until study termination. Catheter status was recorded regarding evidence of partial or complete occlusion or catheter-related infection and any treatment interventions in the interval period between study drug administrations. Termination of the study was marked by catheter removal for occlusion, infection, completion of therapy, or voluntary withdrawal from the study. Study end points were time to first catheter occlusion or time to first catheter-related infection. Complete catheter occlusion was defined as a total occlusion with blockage of at least one catheter lumen marked by an inability to infuse fluid and withdraw blood. Partial occlusion was defined as either an inability to withdraw blood through the device or an inability to flush the device. Catheter-related infection was defined as a positive blood culture drawn from the catheter with no other source of infection. If a blood culture could not be obtained from the device, a catheter-related infection was defined as a positive peripheral culture with no other source of infection or a negative peripheral-blood culture with a clinical response after catheter removal and a positive tip culture. Eighteen months into the planned 3-year timeline, the study was closed without complete accrual because of restrictions imposed by the US Food and Drug Administration on the manufacture and release of urokinase. Differences in proportions were tested for statistical significance using 2 analysis or Fisher s exact test, as appropriate. Occlusions and infections were modeled as a Poisson process stratified by the study medication and device type. Kaplan- Meier estimates of occlusion-free and infection-free rates were used to compare treatments using the log-rank test. Significance was set at P.05. RESULTS Between July 1997 and December 1998, 577 patients from 29 institutions were enrolled onto the study, with a planned accrual of 680 patients. A total of 569 patients were studied. Patient characteristics are presented in Table 1. Two hundred eighty-one patients had port devices, and 288 had external catheters. The median age of the patients was 4.5 years, with a bimodal distribution at 2 years and 15 years. Sixty-three percent of the diagnoses were either leukemia or lymphoma, while solid tumors comprised 35% of study entries. Fifty-one percent of all patients had external catheters, and 86% of these were double-lumen. Forty-nine percent of all patients had ports, and 85% of these were singlelumen. Mean standard deviation duration of indwelling days was similar for both groups: external catheters, days; ports, days (not significant). Median duration on study for the entire population was 4.5 months. Patients with external catheters had a greater number of complications that required device removal compared with port devices (23% v 9%, P.001). Infection was the most common long-term complication that required device removal and occurred in 16 ports (6%) and 43 external catheters (16%; P.001). An occlusive event that led to device removal occurred in a total of 16 patients, with four

3 Dillon et al Table 1. Study Group Characteristics No. Treated % Age group, years Diagnosis Acute lymphocytic leukemia 42 Acute myeloid leukemia 9 CNS lymphoma 8 Tumors of bone 8 Hodgkin s disease 6 Soft-tissue sarcoma 6 Renal 6 Non-Hodgkin s lymphoma 5 Neuroblastoma 4 Other 5 Patients with ports Urokinase treatment 136 Heparin treatment 144 Complications 24 Infection 20 Occlusion 4 Patients with external catheters Urokinase treatment 148 Heparin treatment 140 Complications 65 Infection 53 Occlusion 12 (1.5%) occurring in ports and 12 (4.5%) occurring in external catheters (P.01). Infection and occlusion together that required device removal was reported in a total of seven patients (1.3%; one port and six external catheters). Two hundred eighty-seven patients were randomly assigned to receive urokinase, and 284 were actually treated (136 ports, 148 external catheters). Two hundred ninety patients were randomly assigned to receive heparin, and 284 were treated (144 ports, 140 external catheters). Among all patients, those who received urokinase had fewer occlusive events (partial total) than those who received heparin (23% v 31%, P.02). Time to the first occlusive event was also significantly affected by urokinase administration, as determined by log-rank analysis (P.006; Fig 1). Poisson regression analysis showed a statistically significant 1.6-fold difference in the rate of occlusive events between patients who received urokinase (1.7 events per 1,000 days) and heparin (2.8 events per 1,000 days; P.003). When stratified by type of device, urokinase administration significantly affected the time to the first occlusive event in ports (log-rank, P.03; Fig 2) and demonstrated a 1.7-fold lower overall event rate (1.23 events per 1,000 days) Fig 1. Kaplan-Meier estimate of time to first occlusive event in urokinase and heparin groups for all devices. when compared with heparin administration (2.2 events per 1,000 days; P.03). With external catheters, urokinase also significantly affected the overall event rate, with a 1.5- fold difference between the two study arms (urokinase: 2.3 events per 1,000 days; heparin: 3.7 events per 1,000 days; P.03) with a log-rank value of P.08 for time to first event (Fig 3). No differences in the number of total occlusive events were noted in either the urokinase or heparin treatment groups as a whole or when stratified by device type. The rate of infection for all devices combined was 1.4-fold less in the urokinase group (1.6 events per 1,000 days) than in the heparin group (2.2 events per 1,000 days; P.05), with a log-rank value of P.07 for time to first event (Fig 4). When stratified according to device, only external catheters had evidence of a treatment effect with urokinase (2.6 events per 1,000 days) when compared with heparin (3.9 events per 1,000 days; P.04). The log-rank value was P.10 (Fig 5). In ports, no difference in the rates Fig 2. Kaplan-Meier estimate of time to first occlusive event in urokinase and heparin groups for ports JOURNAL OF CLINICAL ONCOLOGY

4 Urokinase Prophylaxis in Pediatric Vascular Devices Fig 3. Kaplan-Meier estimate of time to first occlusive event in urokinase and heparin groups for external catheters. Fig 5. Kaplan-Meier estimate of time to first infection in urokinase and heparin groups in external catheters. of infection was noted between the two treatment arms of the study (Fig 6). Throughout the study, there were no serious side effects or untoward complications attributed to the administration of either urokinase or heparin. DISCUSSION Infection and occlusion are the most common complications associated with central venous catheters, and lead to the removal of a significant number of devices. Strategies to reduce the incidence of these complications need to be explored in order to maintain access in patients requiring long-term parenteral therapies. Furthermore, the development of such complications imposes additional risks to patients and impacts overall quality of care, particularly if premature removal or replacement of the device is required. Fibrin sheath formation around the external portion of the catheter and within the catheter lumen has been implicated as a major contributing factor in both occlusive and infectious events. 11,12 Interventions designed to decrease fibrin deposition and thrombus formation have the potential to reduce these line-related complications. The findings of this study are consistent with previous reports that a significant proportion of external catheters and implanted ports were affected by thrombotic occlusion and infection, and that external catheters had a significantly higher incidence of these complications than implanted ports Though the median time in our study was slightly longer than 4 months, 23% of patients with external catheters in this study had problems with either occlusions or infections, while such events occurred in only 9% of children with implanted ports. Reasons for the differences in complications between the two types of devices may be attributed to the underlying diagnosis and type of malignancy, intensity of therapy, and frequency with which the devices are accessed. Because this study was randomized by treatment rather than by device, the urokinase and heparin Fig 4. Kaplan-Meier estimate of time to first infection in urokinase and heparin groups for all devices. Fig 6. Kaplan-Meier estimate of time to first infection in urokinase and heparin groups for ports

5 Dillon et al treatment groups were different. A greater percentage of children younger than 2 years and patients with acute myelogenous leukemia, neuroblastoma, bone, and soft tissue sarcomas received external catheters. Stratification by treatment within each device population eliminated any bias of the analysis for each group. Regarding the efficacy of routine fibrinolytic therapy in preventing occlusive or infectious events in vascular access devices in children, this study demonstrated that urokinase administered as a lock every other week can significantly affect the rate at which occlusive events occur in implanted ports and external catheters, and infectious events in external catheters when compared with standard heparin flush administration. The event rates for these complications were similar to those found in other studies of both children and adults The apparent lack of effect of urokinase on infections in implanted ports might have been due in part to the overall lower rate of infection in these devices as compared with external catheters. Though a significant proportion of subjects were not followed up for the intended period due to premature closure of the study, the primary end points of the study were time to first infection or occlusion. Thus, the early closure affected long-term follow-up, but did not significantly compromise the primary study end points. Though thrombolytic therapy is believed to be both safe and effective for the treatment of infected and occluded catheters in both adults and children, there are few studies that have explored the prophylactic use of thrombolytic agents to potentially reduce the incidence of catheter infections and occlusions. Lawson et al suggested the monthly use of urokinase for preventing partial occlusions. 7 A pilot study in adult cancer patients compared a monthly instillation of heparin with urokinase administration in implanted ports. Patients in the urokinase group had lower rates of infection and catheter occlusion compared with those in the heparin group. 17 Similar findings were reported in a population of high-risk adult and pediatric cancer patients with both implanted and external devices who received urokinase every 2 weeks. 17 Ray et al investigated a similar instillation schedule of urokinase in external catheters and reported a reduction in the incidence of thrombotic and infectious complications. 8 Weekly thrombolytic prophylaxis with urokinase in a group of 15 pediatric cancer patients was reported to reduce the rate of central venous catheter related complications as compared with patients who did not receive scheduled thrombolytic treatment. 9 These studies support the hypothesis that the routine administration of thrombolytic therapy into a vascular access device results in a lower rate of infectious and occlusive device-related events. However, other studies have shown contradictory results. A study in children that compared monthly administrations of urokinase and heparin with heparin alone in catheters did not show significant improvement in infections. 18 In addition, Solomon et al also reported lack of efficacy of twice-weekly urokinase infusions in patients with double-lumen external catheters who underwent hematopoietic stem-cell transplantation or intensive chemotherapy for hematologic malignancies. 19 Their protocol used urokinase at a concentration of 2,500 U/mL. Both of those studies differed significantly in the intensity of urokinase administration either with the time schedule or concentration of dose administered compared with our study. It is clear that the effectiveness of thrombolytic therapy in decreasing infectious and thrombotic events is related to the concentration and timing of the dose, the type of venous access device being treated, and perhaps to the intensity of the concomitant chemotherapy program. This clinical trial was terminated, and urokinase removed from the market due to US Food and Drug Administration concerns regarding potential adverse outcomes of established urokinase production procedures, though no adverse outcomes or infection problems related to urokinase administration had ever been or were ever documented. Urokinase has since been reintroduced for clinical use, though not in a formulation typically used in the treatment of device-related occlusions. A recombinant urokinase product is under development. Other fibrinolytic agents such as alteplase and streptokinase are clinically available and have shown efficacy in clearing thrombotic occlusions in catheters. However, there are no published data investigating the ability of these agents to prevent catheter complications similar to this study. Low-dose coumadin therapy has been used to decrease the incidence of catheter-related deep vein thrombosis, but there is no evidence that such therapy decreases the incidence of device infections or occlusions. 20,21 The results of this study must be interpreted with caution, because most patients were treated for 4 months or less, and the actual event rates of infectious and occlusive complications in implanted ports were lower than anticipated. Nonetheless, our study did show that routine urokinase administration could affect the rates at which these device complications occur in children and can serve as a foundation for future long-term studies. Regular administration of thrombolytic therapy in the long-term management of vascular access devices in children undergoing chemotherapy can affect the rate at which occlusive and infectious events occur in these devices when compared with standard heparin flush administration. The routine use of such therapy may decrease the incidence of these complications. Authors Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest JOURNAL OF CLINICAL ONCOLOGY

6 Urokinase Prophylaxis in Pediatric Vascular Devices REFERENCES 1. Wiener ES, McGuire P, Stolar CJH, et al: The CCSG prospective study of venous access devices: An analysis of insertions and causes for removal. J Pediatr Surg 27: , Fan CM: Tunneled catheters. Semin Intervent Radiol 15: , Darouiche RO: Device-associated infections: A macroproblem that starts with microadherence. Clin Infect Dis 33: , Musher D, Goldsmith E, Dunbar S, et al: Association of hypercoagulable states and increased platelet adhesion and aggregation with bacterial colonization of intravenous catheters. J Infect Dis 186: , Lawson M, Bottino JC, Hurtubise MR, et al: The use of urokinase to restore the patency of occluded central venous catheters. Am J Intraven Ther Clin Nutr 9:29-30, Wachs T: Urokinase administration in pediatric patients with occluded central venous catheters. J Intraven Nurs 13: , Lawson M: Partial occlusion of indwelling central venous catheters. J Intraven Nurs 14: , Bagnall-Reeb HA, Saltzman DA, Smith CM, et al: Prophylactic scheduled urokinase for the prevention of right atrial catheter infection. Oncol Nurs Forum 20:315, Ray CE, Shenoy S, McCarthy PL, et al: Weekly prophylactic urokinase installation in tunneled central venous access devices. J Vasc Interv Radiol 10: , Kalmanti M, Germanakis J, Stiakaki E, et al: Prophylaxis with urokinase in pediatric oncology patients with central venous catheters. Pediatr Hematol Oncol 19: , O Farrell L, Griffith JW, Lang CM: Histologic development of the sheath that forms around long-term implanted central venous catheters. JPEN J Parenter Enteral Nutr 20: , Mehall JR, Saltzman DA, Jackson RJ, et al: Fibrin sheath enhances central venous catheter infection. Crit Care Med 30: , Guenier C, Ferreira J, Pector JC: Prolonged venous access in cancer patients. Eur J Surg Oncol 15: , Mirro J Jr, Rao BN, Sokes DC, et al: A prospective study of hickman/broviac catheters and implantable ports in pediatric oncology patients. J Clin Oncol 7: , Severien C, Nelson JD: Frequency of infections associated with implantable systems vs cuffed, tunneled silastic venous catheters in patients with acute leukemia. Am J Dis Child 145: , Groeger JS, Lucas AB, Thaler HT, et al: Infectious morbidity associated with long-term use of venous access devices in patients with cancer. Ann Intern Med 119: , Fraschini G, Becker M, Bruso P, et al: Comparative trial of urokinase versus heparin as prophylaxis for central venous ports. Proc Am Soc Clin Oncol 10:337, 1991 (abstr 1193) 18. Aquino VM, Sandler EX, Mustafa MM, et al: A prospective double blind randomized trial of urokinase flushes to prevent bacteremia resulting from luminal colonization of subcutaneous central venous catheters. J Pediatr Hematol Oncol 24: , Solomon B, Moore J, Arthur C, et al: Lack of efficacy of twice-weekly urokinase in the prevention of complications associated with Hickman catheters: A multicentre randomized comparison of urokinase and heparin. Eur J Cancer 37: , Bern MM, Lokich JJ, Wallach SR, et al: Very low doses of warfarin can prevent thrombosis in central venous catheters. Ann Intern Med 112: , Boraks P, Seale J, Price J, et al: Prevention of central venous catheter associated thrombosis using minidose warfarin in patients with haematological malignancies. Br J Haematol 101: ,