- Email: [email protected]
A multiinstitutional review of central venous line complications: retained intravascular fragments
Journal of Pediatric Surgery (2009) 44, 972–976
www.elsevier.com/locate/jpedsurg
A multiinstitutional review of central venous line complications: retained intravascular fragments Kris Milbrandta , Paul Beaudrya , Ron Andersonb , Sarah Jonesc , Mike Giacomantoniod , David Sigaleta,⁎ a
Division of Paediatric General Surgery, Alberta Children's Hospital, University of Calgary, Calgary, Alberta, Canada T3B 6A8 b Division of Paediatric Oncology, Alberta Children's Hospital, University of Calgary, Calgary Alberta, Canada T3B 6A8 c Division of Paediatric Surgery, Kingston Children's Hospital, Kingston Ontario, Canada K7L 2V7 d Division of Paediatric General Surgery, IWK Children's Hospital, Halifax Nova Scotia, Canada B3H 1A1 Received 9 January 2009; accepted 15 January 2009
Key words: Central lines; Complications; Removal; Fragments; Pediatric
Abstract Background: There have been many reports of complications of central venous lines in children but limited discussion of the specific problem of retained intravascular fragments after attempted removal. We report on a series of 6 patients from 2 tertiary pediatric hospitals that had intravascular segments of long-term central venous lines that could not be removed and so were left in situ. Methods: We conducted a retrospective multiinstitutional review of long-term central venous lines (Broviacs, Port-A-Caths, and Hickmans) removed in the operating room with a focused chart review and prospective follow-up of those patients that had a failed attempt at removal. Results: A total of 299 central venous lines were removed with 6 patients identified as having fragments of lines left behind (2%). The lines had been in place for an average of 37 ± 12 months. The average follow-up period is now 5.4 ± 3.9 years; none of the patients have developed any symptoms, evidence of thrombus, infection, or catheter migration. Conclusion: Given the 2% incidence rate, the issue of managing a stuck long-term central venous line will face most individuals who place these lines. We have demonstrated that simply ligating the catheter and leaving the fragment in place appears to be a safe option with minimal risk to the patient. © 2009 Elsevier Inc. All rights reserved.
Long-term central venous access plays a vital role in the management of pediatric oncology patients, long-term bowel dysfunction, congenital metabolic diseases, and Presented at the 40th Annual CAPS Meeting, August 21-24, 2008, Toronto, Ontario, Canada. ⁎ Corresponding author. Tel.: +1 403 955 2271. E-mail address: [email protected] (D. Sigalet). 0022-3468/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2009.01.033
other conditions in which a reliable vein to access for medications or blood draws is needed [1]. Although the lines have been highly valued by patients and parents, they carry with them inherent risks. Most of these problems occur either at the time of insertion, such as a pneumothorax, or while the line is in place, such as a line infection [2]. Very little has been written or discussed about what to do should a portion of the line becomes stuck within
A multiinstitutional review of central venous line complications the vein [3]. We performed a multiinstitutional review of patients at 2 pediatric tertiary hospitals to identify patients with retained intravascular fragments, we retrospectively reviewed the patient charts for the details of the line use, and we have since prospectively followed the patients for evidence of complications related to the retained portions. The purpose was to determine the incidence of this complication and to document the safety of leaving central line fragments in situ.
1. Materials and methods All coded operative cases for “removal of Broviac line or Hickman” or “removal of Port-A-Cath” were identified through surgical bookings at the Alberta Children's Hospital (ACH), Calgary, Alberta, Canada. These included all types of single or multiple lumen lines removed between January 2003 and December 2007. Broviac and Hickman lines were grouped as simple tunneled lines (TLs), and the remainder lines were Port-A-Caths (PCs). All were from Bard access systems, Salt Lake City, UT. Of these, 3 cases were identified as having incomplete removals, and the charts were reviewed. We excluded cases that occurred outside the operating room, removal of non-TLs, or percutaneous indwelling central catheter (PICC) line removals. Patients were reviewed every 3 to 6 months through the oncology clinic and have had ongoing monitoring with Doppler ultrasound for intravascular thrombi and plain chest radiographs for line migration. At the IWK Health Sciences Center, Halifax, Nova Scotia, Canada, those patients identified in a prior report from 1995 to 2000 had their charts reviewed for the development of complications related to the retained fragments [3]. At the time of the chart review, data obtained included age of patient at insertion, duration line remained until removal, reason for insertion and removal, location, line inserted, and type of line (TL or PC). Follow-up of these patients was obtained through clinic notes in the hospital charts as well as reviewing subsequent imaging after the attempts at removal. Table 1
973
2. Results At ACH, 163 long-term central venous lines were removed in the operating room between January 2003 and December 2007. Most of these were simple TLs (n = 110) and then PCs (n = 53). Of these, 3 patients had incomplete removals with fragments of catheter left in situ. At IWK, 136 lines were removed between 1995 and 2000. Most of these were PCs (n = 129), followed by simple TLs (n = 18). Similarly, 3 of these patients had a line with a fragment left in situ because of difficulties at removal for a total of 6 patients of 299 long-term central venous line removals or a 2% incidence. The average age of insertion of these 6 patients was 42.8 ± 14.8 months (mean ± SD), and the lines had been in place for an average of 37 ± 11 months. One patient had a diagnosis of non–Hodgkin's lymphoma (NHL), whereas the remainder had a diagnosis of acute lymphoblastic leukemia (ALL). The types of lines and sites of insertion are outlined in Table 1. The 3 patients at ACH were reviewed and summarized below. Patient 1 had a very calcified tract, and the line was tied off because the surgeon was concerned about it breaking. The line was stuck at the junction of the catheter and venous entrance point. An attempt at endovascular removal failed. The line was grasped through a transfemoral approach but would not come away from the vein wall and actually resulted in a small embolus sent to the lungs that did not result in any postoperative complications. The line was stable on follow-up chest x-ray (CXR), and the patient has been well for 2 years (Fig. 1). Patient 2 also had a very calcified tract that broke on dissection, and the remainder was left. Postoperative ultrasound showed the catheter to be adherent to vein wall with normal flow and no clot in the vein. Follow-up CXR and ultrasound demonstrated no movement of line at 1 year. A cardiac magnetic resonance imaging was also obtained, which demonstrated the line adherent to the vein wall. Patient 3 was our latest patient, and the line tract was not calcified but adherent to subclavian vein wall. A CXR done in the operating room demonstrated no calcification of the tract, no intravascular knot, and a normal position. It was left
The types of lines and sites of insertion
Patient ID
Age at insertion (mo)
Duration of line (mo)
Type
Location
Diagnosis
Duration of follow-up (y)
Patient 1 Patient 2 Patient 3 a Patient 4 Patient 5 Patient 6 Average
39 38 72 24 36 48 43 ± 16
40 31 36 59 24 32 37 ± 11
Broviac Port Port Port Port Port
Left ext jug Left subclavian Right subclavian Left ext jug Left ext jug Left subclavian
ALL ALL ALL NHL ALL ALL
0.5 1 5 10 8 8 5.4 ± 3.9
The average data show mean ± SD. a Line inserted in Singapore.
974
K. Milbrandt et al.
Fig. 1
Patient 1 showing retained calcified line on the left and grasping of line at attempted endovascular retrieval on the right.
in situ after removing the hub portion of the PC and tying the distal end of the line off. The 3 patients from IWK have been previously summarized [3]. On follow-up, there have been no complications in these patients, including thrombus formation, infection, or migration for now up to 7 years.
3. Discussion There have been several case reports of immovable PICC central lines [4] and broken short-term central lines in adult and pediatric patients [5]. However, the cause and optimal management of immovable central venous lines is unknown. This series is the largest in the English literature, with data from 2 centers combined to gain insight into the incidence and outcome of immovable long-term central venous lines in children. Surprisingly, the problem is not uncommon because we have found an incidence of 2% in a large series of patients. If a catheter appears fixed when removal is attempted, the options are to either leave the catheter fragment in situ or to attempt either intravascular removal or open surgical removal. Potential open techniques carry with them significant risk, necessitating thoracotomy and possible extensive venous reconstruction particularly if the line is placed in the subclavian vein. In this series, an intravascular approach was attempted but failed because the line was adherent to a significant length of the vein wall. This was demonstrated by the fact that the snare could not be placed more proximally than the distal catheter (Fig. 1), implying that the more proximal catheter was adherent to the vein wall. In this case, the fragment was pulled forcibly for 10 to 15 cm, without any evidence of loosening or moving the line. After several attempts, the patient had an episode of unexplained desaturation, which was attributed to an embolic event, and the endovascular
procedure was terminated. Similar findings were reported by Bessoud et al [6], who described several failed attempts at interventional retrieval in their series and related it to either too small of a fragment that had embolized to a small vessel or endothelial formation around the catheter with no free end to grasp. Presumably, fixation of these catheters is caused by the formation of a scar, often with calcification, of the “fibrin sheath” around the catheter. The fibrin sheath found around these indwelling catheters was first described in 1971 [7]. The histology of what happens to long-term central lines has been discussed by several authors. In short-term lines, there is initially an area of endothelial injury with occasional associated thrombus that can be seen. In longterm catheters, vein wall thickening along the length of the catheter and bridging from the vein wall to catheter is later observed. This tissue contains both cellular and acellular components including fibrin, collagen, and later endothelial cells. Interestingly, an endothelial layer develops after
Fig. 2 Cross-sectional detail of a long-term catheter showing fusion between the endothelial layer around the fibrin sheath and the vein wall (permission to reproduce was obtained from Forauer et al [9]).
A multiinstitutional review of central venous line complications 45 days that is indistinguishable from the vein wall [8,9], and most of the catheter length becomes fixed to the vein wall by bridging between the vein wall proper and the neoendothelium of the fibrin sheath (Fig. 2). Other authors have made similar observations when studying the formation of these sheaths [10,11]. The extent of scar formation is reflected by the occurrence of calcification within the sheath as noted in several of our patients and has been noted by others [12]. The fact that this fixation by scar tissue may extend far along the intraluminal aspect of the catheter may explain why it is so resistant to dislodgement, either by pulling externally by the surgeon or internally by an intravascular snare. The cause of scarring causing catheter fixation in these unusual cases is not clear. It may be that after the initial thrombus formation, local inflammatory mediators induce a response by the venous endothelial cells, as an attempt by the body to “seal” the line over along the length of the vein. This may be accentuated in those patients whose catheters remain longer or who receive chemotherapy for an extended period, as in the present cohort. Five of 6 received extended treatment of ALL, which typically included a combination of methotrexate, adriamycin, and vincristine. It may be one of these agents, or the long-term combination causes an increased inflammatory response or calcification of the line. Retrograde flow of these agents around the catheter between the fibrin sheath and the catheter, especially if the tip is partially occluded by fibrin, may also contribute to this reactive scarring. The material from which the line is constructed (polyurethane or silicone) does not appear to affect the rate of line fixation (Jones and Giacomantonio [3]). In the current series of patients with long-term central lines for multiple indications, only the subset receiving chemotherapy was seen to develop line fixation. This contrasts with those patients whose catheters are used strictly for nutrition, suggesting that chemotherapeutic agents may be associated with reactive scar fixation. Optimal management of retained long-term lines is uncertain. However, from these findings, we can safely recommend ligating and leaving catheters in place in cases where an intravascular catheter is adherent and will not move with careful forceful traction at or near the vein entry site. In several other situations such as retained soft tissue fragments near vital organs such as the heart and brain, this has been commonly accepted as a safe alternative. However, one must assume that these foreign bodies have the potential for complications such as migration or thrombus several year or decades later [13-15]. From our experience, the catheters are well fixed to the vein wall at this time so risk of migration is very small. The long-term risk of thrombus formation is unknown but appears to be very low based on existing data from patients with retained pacemaker wires [16]. Because the rate of these complications in pediatric patients with retained line fragments is poorly characterized,
975
we propose to follow such patients with a yearly chest x-ray to detect line migration and Doppler ultrasound to detect any clot formation. An additional safety point could be made to measure the line at the time of insertion and removal to ensure no fragment has been left behind because a calcified tract or sheath is difficult to distinguish from a retained portion of line. In conclusion, we have shown a series of 6 patients from 2 pediatric hospitals with complications relating to removal of long-term indwelling central venous lines necessitating fragments left in situ. To date, no patient has experienced any significant complication from these lines having been left. Risk factors may include duration of line and type of therapy received. A larger study with several centers would be needed to identify these. It would appear safe to leave these fragments in situ for situations where the line is extremely adherent to the vein wall and not free floating in the circulation. The follow-up of these patients should include imaging to assess for line migration or clot formation at a reasonable interval.
References [1] Strovroff M. IV access in children. Pediatr Clin North Am 1998;45(6): 1373-93. [2] Damascell B. Long- term indwelling catheters in 134 patients. AJR Am J Roentgenol 1997;168:1235-9. [3] Jones S, Giacomantonio M. A complication associated with central line removal in the pediatric population: retained fixed catheter fragments. J Pediatr Surg 2003;38(4):594-6. [4] Bautista AB, Ko SH, Sun SC. Retention of percutaneous venous catheter in the newborn: a report of three cases. Am J Perinatol 1995; 12(1):53-4. [5] Fisher RG, Ferreyro R. Evaluation of current techniques for nonsurgical removal of intravascular iatrogenic foreign bodies. AJR Am J Roentgenol 1978;130(3):541-8. [6] Bessoud B, de Baere T, Kuoch V, et al. Experience at a single institution with endovascular treatment of mechanical complications caused by implanted central venous access devices in pediatric and adult patients. AJR Am J Roentgenol 2003;180(2): 527-32. [7] Hoshal Jr VL, Ause RG, Hoskins PA. Fibrin sleeve formation on indwelling subclavian central venous catheters. Arch Surg 1971;102 (4):353-8. [8] Forauer A, Theoharis C. Histologic changes in the human vein wall adjacent to indwelling central venous catheters. J Vasc Interv Radiol 2003;14:1163-8. [9] Forauer A, Theoharis C, Dasika N. Jugular vein catheter placement: histologic features and development of catheter-related (fibrin) sheaths in a swine model. Radiology 2006;240(2): 427-34. [10] Suojanen JN, Brophy DP, Nasser I. Thrombus on indwelling central venous catheters: the histopathology of “fibrin sheaths”. Cardiovasc Intervent Radiol 2000;23(3):194-7. [11] Xiang DZ, Verbeken EK, Van Lommel AT, et al. Intimal hyperplasia after long-term venous catheterization. Eur Surg Res 2000;32(4): 236-45. [12] Andersen MAB, Poenaru D, Kamal I. Calcified catheter “cast”: a rare complication of indwelling central venous catheters in infants. Pediatr Surg Int 1998;13:610-2.
976 [13] Eylon S, Mosheiff R, Liebergall M, et al. Delayed reaction to shrapnel retained in soft tissue. Injury 2005;36(2):275-81. [14] Actis Dato GM, Arslanian A, Di Marzio P, et al. Posttraumatic and iatrogenic foreign bodies in the heart: report of fourteen cases and review of the literature. J Thorac Cardiovasc Surg 2003;126(2): 408-14.
K. Milbrandt et al. [15] Marquardt G, Schick U, Moller-Hartmann W. Brain abscess decades after a penetrating shrapnel injury. Br J Neurosurg 2000; 14(3):246-8. [16] Van Den Akker-Berman LM, Pinzur S, Aydinalp A, et al. Uneventful 25 year course of an intracardiac intravenous catheter fragment in the right heart. J Interv Cardiol 2002;15(5):421-4.
www.elsevier.com/locate/jpedsurg
A multiinstitutional review of central venous line complications: retained intravascular fragments Kris Milbrandta , Paul Beaudrya , Ron Andersonb , Sarah Jonesc , Mike Giacomantoniod , David Sigaleta,⁎ a
Division of Paediatric General Surgery, Alberta Children's Hospital, University of Calgary, Calgary, Alberta, Canada T3B 6A8 b Division of Paediatric Oncology, Alberta Children's Hospital, University of Calgary, Calgary Alberta, Canada T3B 6A8 c Division of Paediatric Surgery, Kingston Children's Hospital, Kingston Ontario, Canada K7L 2V7 d Division of Paediatric General Surgery, IWK Children's Hospital, Halifax Nova Scotia, Canada B3H 1A1 Received 9 January 2009; accepted 15 January 2009
Key words: Central lines; Complications; Removal; Fragments; Pediatric
Abstract Background: There have been many reports of complications of central venous lines in children but limited discussion of the specific problem of retained intravascular fragments after attempted removal. We report on a series of 6 patients from 2 tertiary pediatric hospitals that had intravascular segments of long-term central venous lines that could not be removed and so were left in situ. Methods: We conducted a retrospective multiinstitutional review of long-term central venous lines (Broviacs, Port-A-Caths, and Hickmans) removed in the operating room with a focused chart review and prospective follow-up of those patients that had a failed attempt at removal. Results: A total of 299 central venous lines were removed with 6 patients identified as having fragments of lines left behind (2%). The lines had been in place for an average of 37 ± 12 months. The average follow-up period is now 5.4 ± 3.9 years; none of the patients have developed any symptoms, evidence of thrombus, infection, or catheter migration. Conclusion: Given the 2% incidence rate, the issue of managing a stuck long-term central venous line will face most individuals who place these lines. We have demonstrated that simply ligating the catheter and leaving the fragment in place appears to be a safe option with minimal risk to the patient. © 2009 Elsevier Inc. All rights reserved.
Long-term central venous access plays a vital role in the management of pediatric oncology patients, long-term bowel dysfunction, congenital metabolic diseases, and Presented at the 40th Annual CAPS Meeting, August 21-24, 2008, Toronto, Ontario, Canada. ⁎ Corresponding author. Tel.: +1 403 955 2271. E-mail address: [email protected] (D. Sigalet). 0022-3468/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2009.01.033
other conditions in which a reliable vein to access for medications or blood draws is needed [1]. Although the lines have been highly valued by patients and parents, they carry with them inherent risks. Most of these problems occur either at the time of insertion, such as a pneumothorax, or while the line is in place, such as a line infection [2]. Very little has been written or discussed about what to do should a portion of the line becomes stuck within
A multiinstitutional review of central venous line complications the vein [3]. We performed a multiinstitutional review of patients at 2 pediatric tertiary hospitals to identify patients with retained intravascular fragments, we retrospectively reviewed the patient charts for the details of the line use, and we have since prospectively followed the patients for evidence of complications related to the retained portions. The purpose was to determine the incidence of this complication and to document the safety of leaving central line fragments in situ.
1. Materials and methods All coded operative cases for “removal of Broviac line or Hickman” or “removal of Port-A-Cath” were identified through surgical bookings at the Alberta Children's Hospital (ACH), Calgary, Alberta, Canada. These included all types of single or multiple lumen lines removed between January 2003 and December 2007. Broviac and Hickman lines were grouped as simple tunneled lines (TLs), and the remainder lines were Port-A-Caths (PCs). All were from Bard access systems, Salt Lake City, UT. Of these, 3 cases were identified as having incomplete removals, and the charts were reviewed. We excluded cases that occurred outside the operating room, removal of non-TLs, or percutaneous indwelling central catheter (PICC) line removals. Patients were reviewed every 3 to 6 months through the oncology clinic and have had ongoing monitoring with Doppler ultrasound for intravascular thrombi and plain chest radiographs for line migration. At the IWK Health Sciences Center, Halifax, Nova Scotia, Canada, those patients identified in a prior report from 1995 to 2000 had their charts reviewed for the development of complications related to the retained fragments [3]. At the time of the chart review, data obtained included age of patient at insertion, duration line remained until removal, reason for insertion and removal, location, line inserted, and type of line (TL or PC). Follow-up of these patients was obtained through clinic notes in the hospital charts as well as reviewing subsequent imaging after the attempts at removal. Table 1
973
2. Results At ACH, 163 long-term central venous lines were removed in the operating room between January 2003 and December 2007. Most of these were simple TLs (n = 110) and then PCs (n = 53). Of these, 3 patients had incomplete removals with fragments of catheter left in situ. At IWK, 136 lines were removed between 1995 and 2000. Most of these were PCs (n = 129), followed by simple TLs (n = 18). Similarly, 3 of these patients had a line with a fragment left in situ because of difficulties at removal for a total of 6 patients of 299 long-term central venous line removals or a 2% incidence. The average age of insertion of these 6 patients was 42.8 ± 14.8 months (mean ± SD), and the lines had been in place for an average of 37 ± 11 months. One patient had a diagnosis of non–Hodgkin's lymphoma (NHL), whereas the remainder had a diagnosis of acute lymphoblastic leukemia (ALL). The types of lines and sites of insertion are outlined in Table 1. The 3 patients at ACH were reviewed and summarized below. Patient 1 had a very calcified tract, and the line was tied off because the surgeon was concerned about it breaking. The line was stuck at the junction of the catheter and venous entrance point. An attempt at endovascular removal failed. The line was grasped through a transfemoral approach but would not come away from the vein wall and actually resulted in a small embolus sent to the lungs that did not result in any postoperative complications. The line was stable on follow-up chest x-ray (CXR), and the patient has been well for 2 years (Fig. 1). Patient 2 also had a very calcified tract that broke on dissection, and the remainder was left. Postoperative ultrasound showed the catheter to be adherent to vein wall with normal flow and no clot in the vein. Follow-up CXR and ultrasound demonstrated no movement of line at 1 year. A cardiac magnetic resonance imaging was also obtained, which demonstrated the line adherent to the vein wall. Patient 3 was our latest patient, and the line tract was not calcified but adherent to subclavian vein wall. A CXR done in the operating room demonstrated no calcification of the tract, no intravascular knot, and a normal position. It was left
The types of lines and sites of insertion
Patient ID
Age at insertion (mo)
Duration of line (mo)
Type
Location
Diagnosis
Duration of follow-up (y)
Patient 1 Patient 2 Patient 3 a Patient 4 Patient 5 Patient 6 Average
39 38 72 24 36 48 43 ± 16
40 31 36 59 24 32 37 ± 11
Broviac Port Port Port Port Port
Left ext jug Left subclavian Right subclavian Left ext jug Left ext jug Left subclavian
ALL ALL ALL NHL ALL ALL
0.5 1 5 10 8 8 5.4 ± 3.9
The average data show mean ± SD. a Line inserted in Singapore.
974
K. Milbrandt et al.
Fig. 1
Patient 1 showing retained calcified line on the left and grasping of line at attempted endovascular retrieval on the right.
in situ after removing the hub portion of the PC and tying the distal end of the line off. The 3 patients from IWK have been previously summarized [3]. On follow-up, there have been no complications in these patients, including thrombus formation, infection, or migration for now up to 7 years.
3. Discussion There have been several case reports of immovable PICC central lines [4] and broken short-term central lines in adult and pediatric patients [5]. However, the cause and optimal management of immovable central venous lines is unknown. This series is the largest in the English literature, with data from 2 centers combined to gain insight into the incidence and outcome of immovable long-term central venous lines in children. Surprisingly, the problem is not uncommon because we have found an incidence of 2% in a large series of patients. If a catheter appears fixed when removal is attempted, the options are to either leave the catheter fragment in situ or to attempt either intravascular removal or open surgical removal. Potential open techniques carry with them significant risk, necessitating thoracotomy and possible extensive venous reconstruction particularly if the line is placed in the subclavian vein. In this series, an intravascular approach was attempted but failed because the line was adherent to a significant length of the vein wall. This was demonstrated by the fact that the snare could not be placed more proximally than the distal catheter (Fig. 1), implying that the more proximal catheter was adherent to the vein wall. In this case, the fragment was pulled forcibly for 10 to 15 cm, without any evidence of loosening or moving the line. After several attempts, the patient had an episode of unexplained desaturation, which was attributed to an embolic event, and the endovascular
procedure was terminated. Similar findings were reported by Bessoud et al [6], who described several failed attempts at interventional retrieval in their series and related it to either too small of a fragment that had embolized to a small vessel or endothelial formation around the catheter with no free end to grasp. Presumably, fixation of these catheters is caused by the formation of a scar, often with calcification, of the “fibrin sheath” around the catheter. The fibrin sheath found around these indwelling catheters was first described in 1971 [7]. The histology of what happens to long-term central lines has been discussed by several authors. In short-term lines, there is initially an area of endothelial injury with occasional associated thrombus that can be seen. In longterm catheters, vein wall thickening along the length of the catheter and bridging from the vein wall to catheter is later observed. This tissue contains both cellular and acellular components including fibrin, collagen, and later endothelial cells. Interestingly, an endothelial layer develops after
Fig. 2 Cross-sectional detail of a long-term catheter showing fusion between the endothelial layer around the fibrin sheath and the vein wall (permission to reproduce was obtained from Forauer et al [9]).
A multiinstitutional review of central venous line complications 45 days that is indistinguishable from the vein wall [8,9], and most of the catheter length becomes fixed to the vein wall by bridging between the vein wall proper and the neoendothelium of the fibrin sheath (Fig. 2). Other authors have made similar observations when studying the formation of these sheaths [10,11]. The extent of scar formation is reflected by the occurrence of calcification within the sheath as noted in several of our patients and has been noted by others [12]. The fact that this fixation by scar tissue may extend far along the intraluminal aspect of the catheter may explain why it is so resistant to dislodgement, either by pulling externally by the surgeon or internally by an intravascular snare. The cause of scarring causing catheter fixation in these unusual cases is not clear. It may be that after the initial thrombus formation, local inflammatory mediators induce a response by the venous endothelial cells, as an attempt by the body to “seal” the line over along the length of the vein. This may be accentuated in those patients whose catheters remain longer or who receive chemotherapy for an extended period, as in the present cohort. Five of 6 received extended treatment of ALL, which typically included a combination of methotrexate, adriamycin, and vincristine. It may be one of these agents, or the long-term combination causes an increased inflammatory response or calcification of the line. Retrograde flow of these agents around the catheter between the fibrin sheath and the catheter, especially if the tip is partially occluded by fibrin, may also contribute to this reactive scarring. The material from which the line is constructed (polyurethane or silicone) does not appear to affect the rate of line fixation (Jones and Giacomantonio [3]). In the current series of patients with long-term central lines for multiple indications, only the subset receiving chemotherapy was seen to develop line fixation. This contrasts with those patients whose catheters are used strictly for nutrition, suggesting that chemotherapeutic agents may be associated with reactive scar fixation. Optimal management of retained long-term lines is uncertain. However, from these findings, we can safely recommend ligating and leaving catheters in place in cases where an intravascular catheter is adherent and will not move with careful forceful traction at or near the vein entry site. In several other situations such as retained soft tissue fragments near vital organs such as the heart and brain, this has been commonly accepted as a safe alternative. However, one must assume that these foreign bodies have the potential for complications such as migration or thrombus several year or decades later [13-15]. From our experience, the catheters are well fixed to the vein wall at this time so risk of migration is very small. The long-term risk of thrombus formation is unknown but appears to be very low based on existing data from patients with retained pacemaker wires [16]. Because the rate of these complications in pediatric patients with retained line fragments is poorly characterized,
975
we propose to follow such patients with a yearly chest x-ray to detect line migration and Doppler ultrasound to detect any clot formation. An additional safety point could be made to measure the line at the time of insertion and removal to ensure no fragment has been left behind because a calcified tract or sheath is difficult to distinguish from a retained portion of line. In conclusion, we have shown a series of 6 patients from 2 pediatric hospitals with complications relating to removal of long-term indwelling central venous lines necessitating fragments left in situ. To date, no patient has experienced any significant complication from these lines having been left. Risk factors may include duration of line and type of therapy received. A larger study with several centers would be needed to identify these. It would appear safe to leave these fragments in situ for situations where the line is extremely adherent to the vein wall and not free floating in the circulation. The follow-up of these patients should include imaging to assess for line migration or clot formation at a reasonable interval.
References [1] Strovroff M. IV access in children. Pediatr Clin North Am 1998;45(6): 1373-93. [2] Damascell B. Long- term indwelling catheters in 134 patients. AJR Am J Roentgenol 1997;168:1235-9. [3] Jones S, Giacomantonio M. A complication associated with central line removal in the pediatric population: retained fixed catheter fragments. J Pediatr Surg 2003;38(4):594-6. [4] Bautista AB, Ko SH, Sun SC. Retention of percutaneous venous catheter in the newborn: a report of three cases. Am J Perinatol 1995; 12(1):53-4. [5] Fisher RG, Ferreyro R. Evaluation of current techniques for nonsurgical removal of intravascular iatrogenic foreign bodies. AJR Am J Roentgenol 1978;130(3):541-8. [6] Bessoud B, de Baere T, Kuoch V, et al. Experience at a single institution with endovascular treatment of mechanical complications caused by implanted central venous access devices in pediatric and adult patients. AJR Am J Roentgenol 2003;180(2): 527-32. [7] Hoshal Jr VL, Ause RG, Hoskins PA. Fibrin sleeve formation on indwelling subclavian central venous catheters. Arch Surg 1971;102 (4):353-8. [8] Forauer A, Theoharis C. Histologic changes in the human vein wall adjacent to indwelling central venous catheters. J Vasc Interv Radiol 2003;14:1163-8. [9] Forauer A, Theoharis C, Dasika N. Jugular vein catheter placement: histologic features and development of catheter-related (fibrin) sheaths in a swine model. Radiology 2006;240(2): 427-34. [10] Suojanen JN, Brophy DP, Nasser I. Thrombus on indwelling central venous catheters: the histopathology of “fibrin sheaths”. Cardiovasc Intervent Radiol 2000;23(3):194-7. [11] Xiang DZ, Verbeken EK, Van Lommel AT, et al. Intimal hyperplasia after long-term venous catheterization. Eur Surg Res 2000;32(4): 236-45. [12] Andersen MAB, Poenaru D, Kamal I. Calcified catheter “cast”: a rare complication of indwelling central venous catheters in infants. Pediatr Surg Int 1998;13:610-2.
976 [13] Eylon S, Mosheiff R, Liebergall M, et al. Delayed reaction to shrapnel retained in soft tissue. Injury 2005;36(2):275-81. [14] Actis Dato GM, Arslanian A, Di Marzio P, et al. Posttraumatic and iatrogenic foreign bodies in the heart: report of fourteen cases and review of the literature. J Thorac Cardiovasc Surg 2003;126(2): 408-14.
K. Milbrandt et al. [15] Marquardt G, Schick U, Moller-Hartmann W. Brain abscess decades after a penetrating shrapnel injury. Br J Neurosurg 2000; 14(3):246-8. [16] Van Den Akker-Berman LM, Pinzur S, Aydinalp A, et al. Uneventful 25 year course of an intracardiac intravenous catheter fragment in the right heart. J Interv Cardiol 2002;15(5):421-4.