- Email: [email protected]
Central venous catheters in children with haemophilia
Blood Reviews (2004) 18, 93–100
www.elsevierhealth.com/journals/blre
Central venous catheters in children with haemophilia Rolf Ljung* Departments of Paediatrics and Coagulation Disorders, Lund University, University Hospital, €, Sweden SE-205 02 Malmo
KEYWORDS Haemophilia A; Haemophilia B; Factor VIII; Factor IX; Catheter
Summary Infections, thrombosis and technical problems are the most frequent complications when using implantable central venous access devices in patients with haemophilia. There seem to be two major experiences concerning infections in noninhibitor patients, one is approx. 0.2 infections per 1000 days and the other approx. 1:0ð0:7 1:6Þ=1000 days. Infections are more frequent in inhibitor patients and one can expect approx. one infection per 6–12 months of use. The figures are low for clinically apparent thrombosis in the larger series on record, but routine venograms were not done in most of these series. In studies where this has been done, a high frequency of abnormalities on venograms have been seen in some but not in others. The final decision to use a central line has to be a compromise between the medical goal, the patient’s bleeding tendency, the social situation and the expected risk of complications at the particular haemophilia center. Some of the complications may be reduced by adequate aseptic measures both during implantation and in the subsequent use and clear basic routines for surveillance of the systems and repeated education of the users. c 2003 Elsevier Ltd. All rights reserved.
Introduction
There is a diversity of opinions among treaters concerning the benefits and side-effects of central venous catheters in patients with haemophilia. Some clinicians are in favour while others are more critical due to the frequency of complications. The aim of this paper is to give an overview of the experiences sofar with the use of central venous lines in patients with haemophilia. Haemophilia A and B are X-chromosomal recessive disorders affecting half the sons of carrier females. The diseases are caused by deficient or defective coagulation Factor VIII (FVIII), respectively, Factor IX (FIX) in blood, and depending on the plasma concentration of FVIII/IX clotting activity *
Tel.: +46-40-33-12-86; fax: +46-40-33-62-26. E-mail address: [email protected] (R. Ljung).
(FVIIIC/IXC), haemophilia is classified as severe (FVIII/IXC <1 U/dL), moderate (FVIII/IX 1–4 U/dL) or mild (FVIII/IXC 5–25 U/dL). In the treatment of haemophilia, the most important component is replacement therapy. Treatment with FVIII concentrates has been available since the late 1950s and FIX concentrate since the early 1970s. Treatment of haemophilia A or B with FVIII, respectively, FIX concentrates, irrespective of whether this is done on the event of a bleed or as a prophylactic infusion, requires uncomplicated venous access. If the child is treated on demand the administration of concentrate for a bleed should preferably be done immediately and by the parents at home, a situation in which safe and easy access to a vein is mandatory. If the patient is treated on a prophylactic regimen, infusions are usually given three times per week or every other day in haemophilia A and twice per week or every third day in
0268-960X/$ - see front matter c 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0268-960X(03)00043-2
94
haemophilia B. Ideally primary prophylaxis should be started at an early age and it should be possible for the parents to administer the concentrate at home.1 This may be very difficult using a peripheral vein in a small boy. When starting treatment with factor concentrates irrespective if the treatment regimen is “on-demand” or “prophylaxis”, the first choice of venous access should be a peripheral vein. Introducing a central venous catheter is a risk that has to be balanced against the potential benefits in the individual patient. Medical indications may be poor access to a peripheral vein for the planned therapy especially for the daily injections during immune tolerance induction in patients with inhibitors. A combined medical and social indication may be if a central venous line makes possible home treatment by the parents at an early age. Implantation of a central venous catheter solely on psychological grounds should be discouraged. A child who is simply afraid of venipuncture needs to be helped by other means.
Evaluating the studies on record Several reports have been published with varying experiences of central venous catheters used in patients with haemophilia.2–13 When evaluating previous studies on central venous lines for treatment in haemophilia, several aspects need to be focused on. First, some series include both patients with external catheters and implanted subcutaneous ports.6 It has been clearly shown in a large series of non-haemophilia children that implantable systems have a much lower risk of infections compared to external catheters; 0.7 infections per 1000 patient days for subcutaneous ports vs. 4.3 for external catheters (p < 0:0001).14 The use of external catheters for long term treatment of children with haemophilia is not justified. Secondly, many series include both haemophilia patients with and without inhibitors. It has been shown in many series that the frequency of infection is higher in patients with inhibitors.3;6;11 In one series with a median follow-up time of 30 months, 23% (9/39) of non-inhibitor patients had complications corresponding to 0.23 complications per 1000 patient days. In comparison, 64% (7/11) of inhibitor patients manifested complications.9 Thirdly, some series include patients with HIV infection who may be immunocompromised and thus naturally have a higher risk of infections than the HIV negative patients.6 Series including many HIV-infected persons do not give an accurate risk figure for the, besides haemophilia, healthy patient considering a venous device.
R. Ljung
Another aspect that may be discussed is that many series, although usually with the majority of patients affected with haemophilia, also include other coagulation disorders.4;8 This may be correct but we do not know if experiences from other coagulation disorders are transferable to haemophilia. Central lines is frequently used in paediatric oncology. However, children undergoing treatment with cytostatic drugs have a higher risk for both infections and catheter-related thrombosis.15 Most series with implantable catheters have been using the Port-A-Cath system and almost all experience from larger series in haemophilia patients are with this system. However, some series report also peripheral intravenous access devices (P.A.S. PortsTM , Slim-PortsTM ).8 These seem to be well accepted by the children and parents. In young children it is less threatening to insert a needle in the periphery of the body and they can avoid the visible profile of the port on the chest. However, peripheral ports have been associated with a higher frequency of thrombophlebitis and thrombosis and the average time the patient may benefit from the device is probably shorter.8 The Percusealâ device (Percuseal Medical, Huskvarna, Sweden) is implanted into the subcutaneous tissue, with the top portion protruding from the surface of the skin. The advantage is painless clotting factor application without skin puncture.16 Although interesting since based on a new technique, its usefulness needs to be demonstrated in larger series. This overview will be mainly focused on the Port-A-Cath system since most data from larger series are based on this device. The most frequent complications with implantable ports are infections, thrombosis and various technical problems.
Infections Infections is the most frequent complication when using a central venous line. Table 1 shows some of the larger studies from recent years. Unfortunately it is usually not possible to exclusively study a cohort of haemophilia children who are HIV-negative and furthermore separate inhibitor patients from noninhibitor patients. It is not always clear to the reader if reported “infections” includes both local infections around the port and proved bacteriaemia/ septichaemia. When discussing infections or other complications one should emphasize that many systems have been used for long time before the complication occurred and that the benefits of the system during the period of use justifies the complication.
Central venous catheters in children
95
Table 1 The rate of infection in recent series with haemophilia patients using central venous lines6;8–11 Study
Number of patients (n)
Rate of infection per 1000 patient days
Comment
Blanchette et al.6 Perkins et al.8
19 35
Three patients with inhib., 3 HIV+ 7/32 inhibitor, 2/32 vWD
Ljung et al.3 Sanagostino et al.13 Miller et al.10 McMahon et al.11
53 15 41 58
Tusell (pers. communication 2002)
20 35
0.7 1.2 (central) 0.7 (peripheral device) 0.19 0.3 0.14 1.6 (without inhib) 4.3 (with inhib) 0.28 (prophylaxis) 0.68 (ITI)
If we first consider the patients with inhibitors, it is obvious that the frequency of infections is higher in this subgroup. In the series by Ljung et al.9 45% of the inhibitor patients (5/11) manifested infections after 1–47 months of use (mean 15 months, median 15 months). In the series by McMahon et al.11 23 infections occurred in 18 central venous lines in patients with inhibitors which means one infection/8.5 patient months or 4.3 infections per 1000 patient days which is a considerable higher frequency than in the other study. In a review, van den Berg17 found that in various studies 50–83% of patients with inhibitors can be expected to get an infection. Collins et al. calculated that children with inhibitors develop one infection/8.3 months compared to one/50 months for non-inhibitor patients.18 In the spanish registry of central venous lines, 54% (19/35) of the patients with inhibitors had infections compared to 30% (6/20) of the patients without inhibitors (Tusell et al. pers. communication and WFH Congress 2002). In the same series, the patients with inhibitors had on average a period of 17.5 months free of infections compared to 41.5 months for the ones without inhibitors (Tusell et al. pers. communication and WFH Congress 2002). One reason for the higher frequency of infections in inhibitor patients may be that these patients have small bleedings around the port after an injection that will stimulate bacterial growth in the subcutaneous tissue. Another explanation could be that the inhibitor patient is often on an immune tolerance induction program including one or two injections per day. The non-inhibitor patient on prophylaxis access the port at most every second day. Frequent use should theoretically increase the risk for infections. However, in one study no relation could be found between the number of punctures of the Port-A-Cath and the frequency of infections.9 On the other hand, in that study only
Eleven patients with inhibitors Two inhibitor patients, 13 on prophylaxis Includes external 77/86 devices Port-A-Cath; 37/58 patients haemophilia Port-A-Caths used for prophylaxis/on demand or ITI
the total number of penetrations of the port was considered and presumed to be evenly distributed over time. One could also speculate if the various concentrates given to inhibitor patients, for example heterogeneous “activated prothrombin concentrates”, may result in more fibrin film on the inner and outer catheter surface which may be a nidus for infectious agents. A reasonable conclusion of the different experiences with Port-A-Cath in patients with inhibitors is that one may expect one infection per 6–12 months of regular (several times per week) use. However, these patients need uncomplicated, easy venous access both for the treatment of acute bleeds and for immune tolerance induction programs and the above mentioned risk of infection has to be judged in that context. When discussing non-inhibitor patients, the medical and social indications of a port have to be considered together with the risk of complications. When evaluating the literature, there seem to be two major experiences concerning infections in non-inhibitor patients. Some series report approx. 0.2 infections per 1000 days and other series approx. 1:0ð0:7 1:6Þ=1000 days (Table 1). The series in Table 1 all have a reasonable size and it is usually possible to distinguish the non-inhibitor, non-HIV patient. However, all these series are retrospective. In a small prospective study of 15 patients on whom 13 were on prophylaxis, Sanagostino and coworkers found 0.33 infections/1000 entries to the port.13 What should be considered as an acceptable frequency of infections for this group of patients depends on the situation for the individual patient and the treatment regimen. The child prone to spontaneous bleeds who should start primary prophylactic treatment from the age of one is a greater challenge for venous access than the child on-demand treatment with infrequent bleeds.
96
Central catheters is only an option for those patients in frequent need of an uncomplicated venous access where the benefits have been calculated to be greater than the risks. The indication has to be discussed with the parents, the social situation and the need for home treatment to be taken into the decision making. Most centres probably start with a peripheral vein with the hope that it will be sufficient. In the best of hands, the patient with a Port A Cath, without inhibitor and on regular prophylaxis, will have maximally 1 catheter-related infection every 3–7 year (Table 1). Serious lifethreatening infections seem to be rare, an exemption being a reported case of bacterial endocarditis.19 The most frequent pathogen seem in all series to be Staphylococcus epidermidis or aureus followed by Streptococcus and Klebsiellae species. The use of antibiotics in the preoperative period varies between centres and remains controversial. Most of the experience is from cancer patients and in some studies antibiotics have been of value20 while in others no such correlation was found.21 More important is probably the overall use of antibiotics in the society and thus the frequency of resistance against antibiotics. Of course vigorous education in aseptic techniques and follow-up and reactivation of the education are key-stones in reducing the risk of infection. It has also been speculated that the common use of EMLAâ -anesthetic cream may play a role.8 A reduction of infections was found after the parents were instructed to scrub the cream off by soap and water and not only an ordinary medical swab in order to remove the residual lipid from the skin. This is probably an experience to include in the education of aseptic techniques. In the event of an infection this does not mean that the catheter and implanted device have to be removed. There are several reports in the
R. Ljung
literature that many catheters may be used after the treatment of an infection. Usually this is the case of an early infection around the port or after bacteriaemia/septichaemia with non-resistant bacteria. If infected with multi-resistant bacteria, yeasts or similar agents the catheter has to be removed. It may be advisable not to use the port in the immediate post-operative period to ensure that no infection develops in the area around the port that may be transferred to the inside of the system. There are anecdotal reports of a beneficial effect of the installation of antibiotics in the catheter between injections during treatment of a bacteriaemia. After treatment of bacteriaemia/septichaemia the system has to be checked repeatedly before it is considered free from bacterial contamination. In many cases it has to be removed and it may be advisable not to insert a new system at the same operative procedure.
Thrombosis Infections are the most frequent side-effect when using central venous catheters. During the first 10 years of use of Port-A-Caths in haemophilia, there were few reports of thrombosis and many series did not report any thrombosis despite long-time follow-up (Table 2).7;9;11;22 However, one should be aware that routine venograms were not done in most of these series. In cases where this has been done for various reasons, a small fibrin clot at the tip of the catheter has sometimes been seen, the clinical significance of which is not clear. A clinical sign of such a fibrin clot may be that it is not possible to withdraw, or there is resistance to withdraw blood from the catheter although it is easy to inject. In the series of Medeiros et al.,22 venogram was done in 13/19 patients who had a totally
Table 2 The frequency of thrombosis when using central venous lines in patients with haemophilia Study
Number of patients
Number of thrombosis
Ljung et al.9 Blanchette et al.6 McMahon et al.11 Medeiros et al.22 Miller et al.10 Tusell et al. (spanish registry) Tusell et al. (17 int. paediatric centers) Koerper et al.26 Blanchette et al.23 Journeycahe et al.24
53 19 86 19 41 55 181 11 16 15
0 1 0 0 0 2 7 2 10 8
Central venous catheters in children
97
implantable system in place for >6 months. None of the patients had signs of thrombosis on physical examination. Venography showed that one had a small non-occlusive thrombosis and another had a minimal narrowing at a site of a previous catheter. The authors conclusion was that the prevalence of clinically relevant thrombosis detectable by contrast venography was zero. During the last years there has been a few reports in the literature that suggests that thrombosis may be a more serious problem than previously recognised.23–26 Thus thrombosis is increasingly being a major concern when discussing indwelling catheters. Koerper and coworkers,26 performed venograpy in 11 children without clinical signs of thrombosis who had an implantable device for 1–5 years. Two out of 11 had a total occlusion of the brachiocephalic vein and collateral vessels. In the series by Blanchette and co-workers,23 16 children were examined by venography and ultrasound and thrombosis or signs of thrombosis were found in 10. The most worrying report is the series by Journeycahe and coworkers.24 Fifteen children with a mean age of 9.2, of whom 3 had inhibitor, and with a central venous catheter (tunneled subclavian) for >12 months were studied. 8/15 children had evidence of thrombosis on routine venograms and all 5/15 with clinical problems had abnormal venogram. However, before 48 months of use none of the patients had abnormal venograms, i.e., the risk of thrombosis increases after many years of use. The clinical implication of this study for the patient with a Port-A-Cath is not clear. Tusell et al. reported clinical thrombosis in 2/55 cases (one verified by imaging) in the spanish registry where venograms were done routinely and in 7/181 (6 verified by imaging) cases in a survey of 17 paediatric centres in 9 countries (Tusell et al. pers. communication and WFH Congress 2002).
There are reports in the literature even of thrombosis in inhibitor patients.25 As is the case with infections when using central lines, there seem to be two different experiences concerning thrombosis. In a few rather large series, but without routine venograms, the risk of clinically relevant thrombosis is considered low and routine venograms not indicated. On the other hand in a few smaller series, but with routine venograms, the risk of thrombosis is considerable and routine venogram may be indicated. It is obvious that further, carefully structured, prospective studies are needed in the field. Several points need to be addressed. First, the site of the catheter may be discussed. Is the risk of thrombosis in the subclavial vein similar to the risk in the internal jugular vein? Secondly, the diagnostic method of thrombosis. Doppler ultrasonography seem to be superior to contrast venography to detect thrombosis in the jugular vein and perhaps it is justified to use both methods. Thirdly, the risk of thrombosis may be related to the type of concentrate used. Theoretically an activated prothrombin concentrate may induce higher risk to activate the coagulation system compared to a recombinant single factor. Furthermore, are genetic thrombophilic factors of importance? Ettingshausen and coworkers (2002) suggest that thrombosis in haemophilia has to be considered as a multifactorial disorder.27 Is there any correlation between bacteraemia and thrombosis?
Other complications As shown in Table 3, there are several technical complications that may occur. When a child grows the catheter may be too short with the tip in a smaller vessel than the optimal position in the
Table 3 Complications of implantable central venous catheters other than infections or thrombosis in various studies with an estimate of risk Complication
Frequency
Study
Technical problems (blockade or buckling of catheter, damage from trauma)
3/16 complications Blockade after a mean of 13 months. Catheter splitting/disconnection after a mean of 2 years 1/16 1/19 0/53 3/86 3%
Ljung et al.9 and McMahon et al.11 McMahon et al.11
Short catheter Erosion of skin over port
Ljung et al.9 Blanchette et al.6 Ljung et al.9 McMahon et al.11 Perkins et al.8
98
superior caval vein close to the right atrium. Blockade, buckling and disconnection of the catheter from the port are other problems that may occur. The risk of blockade of the catheter may be diminished by proper education in the use of saline flush after use and the use of a “heparin lock” after use (a few ml heparin 100 U/ml, depending on the system used). If not used the catheter should be flushed and a new heparin lock be administered at least every second month. Another prophylactic measure may be the routine use of installation of urokinase in the catheter (for example 2–3 ml 5000 U/ml for 1–2 h). How often this has to be done depends on the use of the catheter but at least when the injections in the catheter becomes less smooth. Erosions of the skin over the port has occurred in a few reported cases and has to be considered as a serious complication. Inhibitor patients with bleedings and subsequent infection around the port seem to be at particular risk.
The decision to implant a central venous line At the present stage one has to carefully evaluate the indication for a central venous catheter in a child. The final decision to use a central line in a child has to be a compromise between the medical goal, the patient’s bleeding tendency, the social situation and the expected risk of complications at the particular haemophilia center. For many children the use of a central venous line is indispensable in order to get an appropriate treatment. In my opinion, it must be wrong to draw the conclusion from the current literature that central venous catheters are too hazardous to use in haemophilia children. Many series on record have clearly demonstrated the benefits for the child and family with these devices. In prophylactic treatment, a central venous access device is recommended as an option when access to a peripheral vein is difficult or infeasible for the modern frequent treatment of haemophilia from early age. The alternative is usually a suboptimal, less frequent prophylactic treatment using a peripheral vein. However, it is still a difficult decision in the individual child to know if the risks using a central venous line, to enable frequent prophylactic treatment, is lower and justified in comparison with the risk of a serious bleed if the child can not receive optimal prophylaxis due to problem with access to a peripheral vein. It may be advisable to start early treatment by use of a peripheral vein once per week and successively increase the frequency of
R. Ljung
infusions. If this approach does not work and the child has frequent bleeds, it is usually easier for both the parents and the doctor to make a decision to implant a central venous catheter, and also to accept potential future complications of the device. Children who are not on prophylactic treatment may also benefit from a central venous line to enable immediate treatment at home in the event of a bleed. In immune tolerance induction in a small boy, a central line is usually a prerequisite to enable daily or twice daily infusions. The risk of having an inhibitor justifies a greater risk of the proposed treatment. Several measures can be taken to diminish the risk of complications with an implantable venous access system. Most experience exist with the PortA-Cath system and many surgeons prefer to implant the port on the right thoracic wall and insert the catheter via the jugular (mainly internal jugular in small children) veins instead of the subclavian veins. However, there are no systematic studies to support this opinion. It is advisable to substitute with approx. 50 U/kg FVIII/IX to raise the Factor levels to 75–100 U/dL before surgery, a second dose after 6 hours and then maintain haemostatic levels of the factors by daily infusions for at least five days. It is important to introduce adequate aseptic measures both during implantation and in the subsequent use and clear basic routines for surveillance of the systems. Repeated education of the family and other users is probably the most important factor for uncomplicated use of the system25 . It is important that the tip of the catheter is in the right atrium or in the superior vena cava where the flow of blood diminishes the risk compared to when positioned in a smaller vessel. Annual routine X-ray of the position of the catheter tip seems warranted in this context. One should also consider to examine for thrombophilic factors (AT, Protein C and S, FV Leiden and the prothrombin mutation) in the patient as one factor in the decision making of a central venous line27 . Finally, ideally a central venous line should be used for easy venous during the first trying years of treatment and as soon as possible the parents should be educated how to use and gain experience with the use of a peripheral vein. Usually this makes it possible to remove the device after a couple years of use which per se will diminish the risk of complications which increase by time.
The use of Port-A-Caths A survey of the actual use of Port-A-Cath systems in children was done in a survey of 20 centres in 16
Central venous catheters in children European countries.28 In 3/19 centers >50% of the boys under the age of 6 had Port-A-Cath while none had the device in 7/19 centers. A few children at some centers used ports after the age of six years. The treatment regimen is one reason for differences but the most important factor is the tradition and the experience of central venous lines at the center. In summary, a central venous line is many times necessary and unavoidable especially in children. The final decision to use a central line in a child has to be a compromise between the benefits and risks in the individual case.
Practice Points Why use a central venous catheter? – if difficult to access a peripheral vein – to enable frequent injections in order to get an optimal prophylactic treatment – to enable home treatment at an early age – to enable once or twice daily injections for immunotolerance induction in patients with inhibitors. How to diminish the risk of complications with central venous lines? – Screen for thrombophilic factors – Repeated education of parents – Surveillance routines (saline flush, heparin, urokinase/rtPA) – Regular check-up of position of catheter – Routine venograms after several years of use? – Remove the systems after 4–5 years of use?
Research Agenda Does thrombophilic factors increase the risk for thrombosis in haemophilia children with central venous access devices?
Acknowledgements The swedish results presented in this review was supported by grants from the Swedish Medical Research Council (no. 13493), the University of Lund € University Hospital. (ALF) and funds of Malmo
References 1. Ljung RC. Can haemophilic arthropathy be prevented? Br J Haematol 1998;101:215–9. 2. McWhirter WR, Gray L. Indwelling intravenous catheter in a young haemophiliac. Lancet 1988;2:99–100, letter.
99
3. Ljung R, Petrini P, Lindgren AK, Berntorp E. Implantable central venous catheter facilitates prophylactic treatment in children with haemophilia. Acta Paediatr 1992;81: 918–20. 4. Liesner RJ, Vora AJ, Hann IM, Lilleymann JS. Use of central venous catheters in children with severe congenital coagulopathy. Br J Haematol 1995;91:203–7. 5. Warrier I, Baird-Cox K, Lusher J. Use of central venous catheters in children with Haemophilia: one haemophilia treatment centre experience. Haemophilia 1997;3. 6. Blanchette VS, Al Musa A, Stain AM, Filler RM, Ingram J. Central venous access catheters in children with haemophilia. Blood Coagul Fibrinol 1996;7(Suppl. 1):S39–44. 7. Blanchette VS, Al Musa A, Stain AM, Ingram J, Fille RM. Central venous access devices in children with hemophilia: an update. Blood Coagul Fibrinol 1997;8(Suppl 1):S11–14. 8. Perkins JL, Johnson VA, Osip JM, Christie BA, Nelson SC, Moertel CL, et al. The use of implantable venous access devices (IVADs) in children with hemophilia. J Pediatr Hematol Oncol 1997;19:339–44. 9. Ljung R, van den Berg M, Petrini P, Tengborn L, Scheibel E, Kekomaki R, et al. Port-A-Cath usage in children with haemophilia: experience of 53 cases. Acta Paediatr 1998; 87:1051–4. 10. Miller K, Buchanan GR, Zappa S, Cochran C, Laufenberg J, Medeiros D, et al. Implantable venous access devices in children with hemophilia: a report of low infection rates. J Pediatr 1998;132:934–8, see comments. 11. McMahon C, Smith J, Khair K, Liesner R, Hann I, Smith O. Central venous access devices in children with congenital coagulation disorders; complications and long-term outcome. Brit J Haematol 2000;110:461–8. 12. Mumtaz H, Williams V, Hauer-Jensen M, Rowe M, HenryTillman RS, Heaton K, et al. Central venous catheter placement in patients with disorders of haemostasis. Am J Surg 2001;180:503–6. 13. Santagostino E, Gringeri A, Muca Perja M, Mannucci PM. A prospective clinical trial of implantable central venous access in children with haemophilia. Br J Haematol 1998; 102:1224–8. 14. Ingram J, Wietzman S, Greenberg ML. Complications of indwelling venous access lines in the pediatric hematology patient: a prospective comparison of external venous catheters and subcutaneous ports. J Pediatr Hematol Oncol 1991;13:130–6. 15. Wesenberg F, Flaaten H, Janssen CJ. Central venous catheter with subcutaneous port (Port-A-Cath): 8 years clinical follow-up with children. Paediatr Haematol Oncol 1993;10. €tzsch T, Carlsson M, Lethagen S. The 16. Berntorp E, Frick K, Ma Percuseal device – a new option for effective continuous prophylaxis? Haemophilia 1998;4(3):184. 17. Van Den Berg HM, Fischer K, Roosendaal G, Mauser Bunschoten EP. The use of the Port-A-Cath in children with haemophilia–a review. Haemophilia 1998;4:418–20. 18. Collins PW, Khair KS, Liesner R, Hann IM. Complications experienced with central venous catheters in children with congenital bleeding disorders. Br J Haematol 1997;99: 206–8. 19. Hothi D, Kelsall W, Baglin T, Williams D. Bacterial endocarditis in a child with haemophilia B. risks of central venous catheters. Haemophilia 2001;7. 20. Lim SH, Smith MP, Machin SJ, Goldstone AH. A prospective randomised study of prophylactic teicoplanin to prevent early Hickman catheter-related sepsis in patients receiving intensive chemotherapy for hematological malignancies. Eur J Haematol 1993;(Suppl. 54):10–3.
100
21. Ranson MR, Oppenheim BA, Jackson A, Kamthan AG, Scarffe JH. Double-blind placebo controlled study of vancomycin prophylaxis for central venous catheter insertion in cancer patients. J Hosp Infect 1990;15:95–102. 22. Medeiros D, Miller KL, Rollins NK, Buchanan GR. Contrast venography in young haemophiliacs with implantable central venous access devices. Haemophilia 1998;4: 10–5. 23. Blanchette VS, Al Trabolsi H, Stain AM, Sparling CR, Massicotte MP, Daneman A. High risk of central venous line-associated thrombosis in boys with haemophilia (abstract). Blood 1999;94:234a. 24. Journeycahe JM, Quinn CT, Miller KL, Zajac JC, Buchanan G. Catheter related deep venous thrombosis in children with haemophilia. Blood 2001;98:1717–31.
R. Ljung
25. Vidler V, Richards M, Vora A. Central venous catheterassociated thrombosis in severe haemophilia. Br J Haematol 1999;104:461–4. 26. Koerper MA, Esker S, Cobb L. Asymptomatic thrombosis of innominate vein in haemophilic children with subcutaneous venous access devices (ports) (Abstract). National Haemophilia Foundation 48th Annual Meeting, 1996. San Diego. 27. Ettingshausen C, Kurnik K, Schobess R, Kreuz W, Halimek S, Pollman H, et al. Catheter related thrombosis in children with haemophilia A. evidence of a multifactorial disease. Blood 2002;99:449–500. 28. Ljung R, Aronis-Vournas S, Kurnik-Auberger K, van den Berg M, Chambost H, Claeyssens S, et al. Treatment of children with haemophilia in Europe: a survey of 20 centres in 16 countries. Haemophilia 2000;6:619–24.
www.elsevierhealth.com/journals/blre
Central venous catheters in children with haemophilia Rolf Ljung* Departments of Paediatrics and Coagulation Disorders, Lund University, University Hospital, €, Sweden SE-205 02 Malmo
KEYWORDS Haemophilia A; Haemophilia B; Factor VIII; Factor IX; Catheter
Summary Infections, thrombosis and technical problems are the most frequent complications when using implantable central venous access devices in patients with haemophilia. There seem to be two major experiences concerning infections in noninhibitor patients, one is approx. 0.2 infections per 1000 days and the other approx. 1:0ð0:7 1:6Þ=1000 days. Infections are more frequent in inhibitor patients and one can expect approx. one infection per 6–12 months of use. The figures are low for clinically apparent thrombosis in the larger series on record, but routine venograms were not done in most of these series. In studies where this has been done, a high frequency of abnormalities on venograms have been seen in some but not in others. The final decision to use a central line has to be a compromise between the medical goal, the patient’s bleeding tendency, the social situation and the expected risk of complications at the particular haemophilia center. Some of the complications may be reduced by adequate aseptic measures both during implantation and in the subsequent use and clear basic routines for surveillance of the systems and repeated education of the users. c 2003 Elsevier Ltd. All rights reserved.
Introduction
There is a diversity of opinions among treaters concerning the benefits and side-effects of central venous catheters in patients with haemophilia. Some clinicians are in favour while others are more critical due to the frequency of complications. The aim of this paper is to give an overview of the experiences sofar with the use of central venous lines in patients with haemophilia. Haemophilia A and B are X-chromosomal recessive disorders affecting half the sons of carrier females. The diseases are caused by deficient or defective coagulation Factor VIII (FVIII), respectively, Factor IX (FIX) in blood, and depending on the plasma concentration of FVIII/IX clotting activity *
Tel.: +46-40-33-12-86; fax: +46-40-33-62-26. E-mail address: [email protected] (R. Ljung).
(FVIIIC/IXC), haemophilia is classified as severe (FVIII/IXC <1 U/dL), moderate (FVIII/IX 1–4 U/dL) or mild (FVIII/IXC 5–25 U/dL). In the treatment of haemophilia, the most important component is replacement therapy. Treatment with FVIII concentrates has been available since the late 1950s and FIX concentrate since the early 1970s. Treatment of haemophilia A or B with FVIII, respectively, FIX concentrates, irrespective of whether this is done on the event of a bleed or as a prophylactic infusion, requires uncomplicated venous access. If the child is treated on demand the administration of concentrate for a bleed should preferably be done immediately and by the parents at home, a situation in which safe and easy access to a vein is mandatory. If the patient is treated on a prophylactic regimen, infusions are usually given three times per week or every other day in haemophilia A and twice per week or every third day in
0268-960X/$ - see front matter c 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0268-960X(03)00043-2
94
haemophilia B. Ideally primary prophylaxis should be started at an early age and it should be possible for the parents to administer the concentrate at home.1 This may be very difficult using a peripheral vein in a small boy. When starting treatment with factor concentrates irrespective if the treatment regimen is “on-demand” or “prophylaxis”, the first choice of venous access should be a peripheral vein. Introducing a central venous catheter is a risk that has to be balanced against the potential benefits in the individual patient. Medical indications may be poor access to a peripheral vein for the planned therapy especially for the daily injections during immune tolerance induction in patients with inhibitors. A combined medical and social indication may be if a central venous line makes possible home treatment by the parents at an early age. Implantation of a central venous catheter solely on psychological grounds should be discouraged. A child who is simply afraid of venipuncture needs to be helped by other means.
Evaluating the studies on record Several reports have been published with varying experiences of central venous catheters used in patients with haemophilia.2–13 When evaluating previous studies on central venous lines for treatment in haemophilia, several aspects need to be focused on. First, some series include both patients with external catheters and implanted subcutaneous ports.6 It has been clearly shown in a large series of non-haemophilia children that implantable systems have a much lower risk of infections compared to external catheters; 0.7 infections per 1000 patient days for subcutaneous ports vs. 4.3 for external catheters (p < 0:0001).14 The use of external catheters for long term treatment of children with haemophilia is not justified. Secondly, many series include both haemophilia patients with and without inhibitors. It has been shown in many series that the frequency of infection is higher in patients with inhibitors.3;6;11 In one series with a median follow-up time of 30 months, 23% (9/39) of non-inhibitor patients had complications corresponding to 0.23 complications per 1000 patient days. In comparison, 64% (7/11) of inhibitor patients manifested complications.9 Thirdly, some series include patients with HIV infection who may be immunocompromised and thus naturally have a higher risk of infections than the HIV negative patients.6 Series including many HIV-infected persons do not give an accurate risk figure for the, besides haemophilia, healthy patient considering a venous device.
R. Ljung
Another aspect that may be discussed is that many series, although usually with the majority of patients affected with haemophilia, also include other coagulation disorders.4;8 This may be correct but we do not know if experiences from other coagulation disorders are transferable to haemophilia. Central lines is frequently used in paediatric oncology. However, children undergoing treatment with cytostatic drugs have a higher risk for both infections and catheter-related thrombosis.15 Most series with implantable catheters have been using the Port-A-Cath system and almost all experience from larger series in haemophilia patients are with this system. However, some series report also peripheral intravenous access devices (P.A.S. PortsTM , Slim-PortsTM ).8 These seem to be well accepted by the children and parents. In young children it is less threatening to insert a needle in the periphery of the body and they can avoid the visible profile of the port on the chest. However, peripheral ports have been associated with a higher frequency of thrombophlebitis and thrombosis and the average time the patient may benefit from the device is probably shorter.8 The Percusealâ device (Percuseal Medical, Huskvarna, Sweden) is implanted into the subcutaneous tissue, with the top portion protruding from the surface of the skin. The advantage is painless clotting factor application without skin puncture.16 Although interesting since based on a new technique, its usefulness needs to be demonstrated in larger series. This overview will be mainly focused on the Port-A-Cath system since most data from larger series are based on this device. The most frequent complications with implantable ports are infections, thrombosis and various technical problems.
Infections Infections is the most frequent complication when using a central venous line. Table 1 shows some of the larger studies from recent years. Unfortunately it is usually not possible to exclusively study a cohort of haemophilia children who are HIV-negative and furthermore separate inhibitor patients from noninhibitor patients. It is not always clear to the reader if reported “infections” includes both local infections around the port and proved bacteriaemia/ septichaemia. When discussing infections or other complications one should emphasize that many systems have been used for long time before the complication occurred and that the benefits of the system during the period of use justifies the complication.
Central venous catheters in children
95
Table 1 The rate of infection in recent series with haemophilia patients using central venous lines6;8–11 Study
Number of patients (n)
Rate of infection per 1000 patient days
Comment
Blanchette et al.6 Perkins et al.8
19 35
Three patients with inhib., 3 HIV+ 7/32 inhibitor, 2/32 vWD
Ljung et al.3 Sanagostino et al.13 Miller et al.10 McMahon et al.11
53 15 41 58
Tusell (pers. communication 2002)
20 35
0.7 1.2 (central) 0.7 (peripheral device) 0.19 0.3 0.14 1.6 (without inhib) 4.3 (with inhib) 0.28 (prophylaxis) 0.68 (ITI)
If we first consider the patients with inhibitors, it is obvious that the frequency of infections is higher in this subgroup. In the series by Ljung et al.9 45% of the inhibitor patients (5/11) manifested infections after 1–47 months of use (mean 15 months, median 15 months). In the series by McMahon et al.11 23 infections occurred in 18 central venous lines in patients with inhibitors which means one infection/8.5 patient months or 4.3 infections per 1000 patient days which is a considerable higher frequency than in the other study. In a review, van den Berg17 found that in various studies 50–83% of patients with inhibitors can be expected to get an infection. Collins et al. calculated that children with inhibitors develop one infection/8.3 months compared to one/50 months for non-inhibitor patients.18 In the spanish registry of central venous lines, 54% (19/35) of the patients with inhibitors had infections compared to 30% (6/20) of the patients without inhibitors (Tusell et al. pers. communication and WFH Congress 2002). In the same series, the patients with inhibitors had on average a period of 17.5 months free of infections compared to 41.5 months for the ones without inhibitors (Tusell et al. pers. communication and WFH Congress 2002). One reason for the higher frequency of infections in inhibitor patients may be that these patients have small bleedings around the port after an injection that will stimulate bacterial growth in the subcutaneous tissue. Another explanation could be that the inhibitor patient is often on an immune tolerance induction program including one or two injections per day. The non-inhibitor patient on prophylaxis access the port at most every second day. Frequent use should theoretically increase the risk for infections. However, in one study no relation could be found between the number of punctures of the Port-A-Cath and the frequency of infections.9 On the other hand, in that study only
Eleven patients with inhibitors Two inhibitor patients, 13 on prophylaxis Includes external 77/86 devices Port-A-Cath; 37/58 patients haemophilia Port-A-Caths used for prophylaxis/on demand or ITI
the total number of penetrations of the port was considered and presumed to be evenly distributed over time. One could also speculate if the various concentrates given to inhibitor patients, for example heterogeneous “activated prothrombin concentrates”, may result in more fibrin film on the inner and outer catheter surface which may be a nidus for infectious agents. A reasonable conclusion of the different experiences with Port-A-Cath in patients with inhibitors is that one may expect one infection per 6–12 months of regular (several times per week) use. However, these patients need uncomplicated, easy venous access both for the treatment of acute bleeds and for immune tolerance induction programs and the above mentioned risk of infection has to be judged in that context. When discussing non-inhibitor patients, the medical and social indications of a port have to be considered together with the risk of complications. When evaluating the literature, there seem to be two major experiences concerning infections in non-inhibitor patients. Some series report approx. 0.2 infections per 1000 days and other series approx. 1:0ð0:7 1:6Þ=1000 days (Table 1). The series in Table 1 all have a reasonable size and it is usually possible to distinguish the non-inhibitor, non-HIV patient. However, all these series are retrospective. In a small prospective study of 15 patients on whom 13 were on prophylaxis, Sanagostino and coworkers found 0.33 infections/1000 entries to the port.13 What should be considered as an acceptable frequency of infections for this group of patients depends on the situation for the individual patient and the treatment regimen. The child prone to spontaneous bleeds who should start primary prophylactic treatment from the age of one is a greater challenge for venous access than the child on-demand treatment with infrequent bleeds.
96
Central catheters is only an option for those patients in frequent need of an uncomplicated venous access where the benefits have been calculated to be greater than the risks. The indication has to be discussed with the parents, the social situation and the need for home treatment to be taken into the decision making. Most centres probably start with a peripheral vein with the hope that it will be sufficient. In the best of hands, the patient with a Port A Cath, without inhibitor and on regular prophylaxis, will have maximally 1 catheter-related infection every 3–7 year (Table 1). Serious lifethreatening infections seem to be rare, an exemption being a reported case of bacterial endocarditis.19 The most frequent pathogen seem in all series to be Staphylococcus epidermidis or aureus followed by Streptococcus and Klebsiellae species. The use of antibiotics in the preoperative period varies between centres and remains controversial. Most of the experience is from cancer patients and in some studies antibiotics have been of value20 while in others no such correlation was found.21 More important is probably the overall use of antibiotics in the society and thus the frequency of resistance against antibiotics. Of course vigorous education in aseptic techniques and follow-up and reactivation of the education are key-stones in reducing the risk of infection. It has also been speculated that the common use of EMLAâ -anesthetic cream may play a role.8 A reduction of infections was found after the parents were instructed to scrub the cream off by soap and water and not only an ordinary medical swab in order to remove the residual lipid from the skin. This is probably an experience to include in the education of aseptic techniques. In the event of an infection this does not mean that the catheter and implanted device have to be removed. There are several reports in the
R. Ljung
literature that many catheters may be used after the treatment of an infection. Usually this is the case of an early infection around the port or after bacteriaemia/septichaemia with non-resistant bacteria. If infected with multi-resistant bacteria, yeasts or similar agents the catheter has to be removed. It may be advisable not to use the port in the immediate post-operative period to ensure that no infection develops in the area around the port that may be transferred to the inside of the system. There are anecdotal reports of a beneficial effect of the installation of antibiotics in the catheter between injections during treatment of a bacteriaemia. After treatment of bacteriaemia/septichaemia the system has to be checked repeatedly before it is considered free from bacterial contamination. In many cases it has to be removed and it may be advisable not to insert a new system at the same operative procedure.
Thrombosis Infections are the most frequent side-effect when using central venous catheters. During the first 10 years of use of Port-A-Caths in haemophilia, there were few reports of thrombosis and many series did not report any thrombosis despite long-time follow-up (Table 2).7;9;11;22 However, one should be aware that routine venograms were not done in most of these series. In cases where this has been done for various reasons, a small fibrin clot at the tip of the catheter has sometimes been seen, the clinical significance of which is not clear. A clinical sign of such a fibrin clot may be that it is not possible to withdraw, or there is resistance to withdraw blood from the catheter although it is easy to inject. In the series of Medeiros et al.,22 venogram was done in 13/19 patients who had a totally
Table 2 The frequency of thrombosis when using central venous lines in patients with haemophilia Study
Number of patients
Number of thrombosis
Ljung et al.9 Blanchette et al.6 McMahon et al.11 Medeiros et al.22 Miller et al.10 Tusell et al. (spanish registry) Tusell et al. (17 int. paediatric centers) Koerper et al.26 Blanchette et al.23 Journeycahe et al.24
53 19 86 19 41 55 181 11 16 15
0 1 0 0 0 2 7 2 10 8
Central venous catheters in children
97
implantable system in place for >6 months. None of the patients had signs of thrombosis on physical examination. Venography showed that one had a small non-occlusive thrombosis and another had a minimal narrowing at a site of a previous catheter. The authors conclusion was that the prevalence of clinically relevant thrombosis detectable by contrast venography was zero. During the last years there has been a few reports in the literature that suggests that thrombosis may be a more serious problem than previously recognised.23–26 Thus thrombosis is increasingly being a major concern when discussing indwelling catheters. Koerper and coworkers,26 performed venograpy in 11 children without clinical signs of thrombosis who had an implantable device for 1–5 years. Two out of 11 had a total occlusion of the brachiocephalic vein and collateral vessels. In the series by Blanchette and co-workers,23 16 children were examined by venography and ultrasound and thrombosis or signs of thrombosis were found in 10. The most worrying report is the series by Journeycahe and coworkers.24 Fifteen children with a mean age of 9.2, of whom 3 had inhibitor, and with a central venous catheter (tunneled subclavian) for >12 months were studied. 8/15 children had evidence of thrombosis on routine venograms and all 5/15 with clinical problems had abnormal venogram. However, before 48 months of use none of the patients had abnormal venograms, i.e., the risk of thrombosis increases after many years of use. The clinical implication of this study for the patient with a Port-A-Cath is not clear. Tusell et al. reported clinical thrombosis in 2/55 cases (one verified by imaging) in the spanish registry where venograms were done routinely and in 7/181 (6 verified by imaging) cases in a survey of 17 paediatric centres in 9 countries (Tusell et al. pers. communication and WFH Congress 2002).
There are reports in the literature even of thrombosis in inhibitor patients.25 As is the case with infections when using central lines, there seem to be two different experiences concerning thrombosis. In a few rather large series, but without routine venograms, the risk of clinically relevant thrombosis is considered low and routine venograms not indicated. On the other hand in a few smaller series, but with routine venograms, the risk of thrombosis is considerable and routine venogram may be indicated. It is obvious that further, carefully structured, prospective studies are needed in the field. Several points need to be addressed. First, the site of the catheter may be discussed. Is the risk of thrombosis in the subclavial vein similar to the risk in the internal jugular vein? Secondly, the diagnostic method of thrombosis. Doppler ultrasonography seem to be superior to contrast venography to detect thrombosis in the jugular vein and perhaps it is justified to use both methods. Thirdly, the risk of thrombosis may be related to the type of concentrate used. Theoretically an activated prothrombin concentrate may induce higher risk to activate the coagulation system compared to a recombinant single factor. Furthermore, are genetic thrombophilic factors of importance? Ettingshausen and coworkers (2002) suggest that thrombosis in haemophilia has to be considered as a multifactorial disorder.27 Is there any correlation between bacteraemia and thrombosis?
Other complications As shown in Table 3, there are several technical complications that may occur. When a child grows the catheter may be too short with the tip in a smaller vessel than the optimal position in the
Table 3 Complications of implantable central venous catheters other than infections or thrombosis in various studies with an estimate of risk Complication
Frequency
Study
Technical problems (blockade or buckling of catheter, damage from trauma)
3/16 complications Blockade after a mean of 13 months. Catheter splitting/disconnection after a mean of 2 years 1/16 1/19 0/53 3/86 3%
Ljung et al.9 and McMahon et al.11 McMahon et al.11
Short catheter Erosion of skin over port
Ljung et al.9 Blanchette et al.6 Ljung et al.9 McMahon et al.11 Perkins et al.8
98
superior caval vein close to the right atrium. Blockade, buckling and disconnection of the catheter from the port are other problems that may occur. The risk of blockade of the catheter may be diminished by proper education in the use of saline flush after use and the use of a “heparin lock” after use (a few ml heparin 100 U/ml, depending on the system used). If not used the catheter should be flushed and a new heparin lock be administered at least every second month. Another prophylactic measure may be the routine use of installation of urokinase in the catheter (for example 2–3 ml 5000 U/ml for 1–2 h). How often this has to be done depends on the use of the catheter but at least when the injections in the catheter becomes less smooth. Erosions of the skin over the port has occurred in a few reported cases and has to be considered as a serious complication. Inhibitor patients with bleedings and subsequent infection around the port seem to be at particular risk.
The decision to implant a central venous line At the present stage one has to carefully evaluate the indication for a central venous catheter in a child. The final decision to use a central line in a child has to be a compromise between the medical goal, the patient’s bleeding tendency, the social situation and the expected risk of complications at the particular haemophilia center. For many children the use of a central venous line is indispensable in order to get an appropriate treatment. In my opinion, it must be wrong to draw the conclusion from the current literature that central venous catheters are too hazardous to use in haemophilia children. Many series on record have clearly demonstrated the benefits for the child and family with these devices. In prophylactic treatment, a central venous access device is recommended as an option when access to a peripheral vein is difficult or infeasible for the modern frequent treatment of haemophilia from early age. The alternative is usually a suboptimal, less frequent prophylactic treatment using a peripheral vein. However, it is still a difficult decision in the individual child to know if the risks using a central venous line, to enable frequent prophylactic treatment, is lower and justified in comparison with the risk of a serious bleed if the child can not receive optimal prophylaxis due to problem with access to a peripheral vein. It may be advisable to start early treatment by use of a peripheral vein once per week and successively increase the frequency of
R. Ljung
infusions. If this approach does not work and the child has frequent bleeds, it is usually easier for both the parents and the doctor to make a decision to implant a central venous catheter, and also to accept potential future complications of the device. Children who are not on prophylactic treatment may also benefit from a central venous line to enable immediate treatment at home in the event of a bleed. In immune tolerance induction in a small boy, a central line is usually a prerequisite to enable daily or twice daily infusions. The risk of having an inhibitor justifies a greater risk of the proposed treatment. Several measures can be taken to diminish the risk of complications with an implantable venous access system. Most experience exist with the PortA-Cath system and many surgeons prefer to implant the port on the right thoracic wall and insert the catheter via the jugular (mainly internal jugular in small children) veins instead of the subclavian veins. However, there are no systematic studies to support this opinion. It is advisable to substitute with approx. 50 U/kg FVIII/IX to raise the Factor levels to 75–100 U/dL before surgery, a second dose after 6 hours and then maintain haemostatic levels of the factors by daily infusions for at least five days. It is important to introduce adequate aseptic measures both during implantation and in the subsequent use and clear basic routines for surveillance of the systems. Repeated education of the family and other users is probably the most important factor for uncomplicated use of the system25 . It is important that the tip of the catheter is in the right atrium or in the superior vena cava where the flow of blood diminishes the risk compared to when positioned in a smaller vessel. Annual routine X-ray of the position of the catheter tip seems warranted in this context. One should also consider to examine for thrombophilic factors (AT, Protein C and S, FV Leiden and the prothrombin mutation) in the patient as one factor in the decision making of a central venous line27 . Finally, ideally a central venous line should be used for easy venous during the first trying years of treatment and as soon as possible the parents should be educated how to use and gain experience with the use of a peripheral vein. Usually this makes it possible to remove the device after a couple years of use which per se will diminish the risk of complications which increase by time.
The use of Port-A-Caths A survey of the actual use of Port-A-Cath systems in children was done in a survey of 20 centres in 16
Central venous catheters in children European countries.28 In 3/19 centers >50% of the boys under the age of 6 had Port-A-Cath while none had the device in 7/19 centers. A few children at some centers used ports after the age of six years. The treatment regimen is one reason for differences but the most important factor is the tradition and the experience of central venous lines at the center. In summary, a central venous line is many times necessary and unavoidable especially in children. The final decision to use a central line in a child has to be a compromise between the benefits and risks in the individual case.
Practice Points Why use a central venous catheter? – if difficult to access a peripheral vein – to enable frequent injections in order to get an optimal prophylactic treatment – to enable home treatment at an early age – to enable once or twice daily injections for immunotolerance induction in patients with inhibitors. How to diminish the risk of complications with central venous lines? – Screen for thrombophilic factors – Repeated education of parents – Surveillance routines (saline flush, heparin, urokinase/rtPA) – Regular check-up of position of catheter – Routine venograms after several years of use? – Remove the systems after 4–5 years of use?
Research Agenda Does thrombophilic factors increase the risk for thrombosis in haemophilia children with central venous access devices?
Acknowledgements The swedish results presented in this review was supported by grants from the Swedish Medical Research Council (no. 13493), the University of Lund € University Hospital. (ALF) and funds of Malmo
References 1. Ljung RC. Can haemophilic arthropathy be prevented? Br J Haematol 1998;101:215–9. 2. McWhirter WR, Gray L. Indwelling intravenous catheter in a young haemophiliac. Lancet 1988;2:99–100, letter.
99
3. Ljung R, Petrini P, Lindgren AK, Berntorp E. Implantable central venous catheter facilitates prophylactic treatment in children with haemophilia. Acta Paediatr 1992;81: 918–20. 4. Liesner RJ, Vora AJ, Hann IM, Lilleymann JS. Use of central venous catheters in children with severe congenital coagulopathy. Br J Haematol 1995;91:203–7. 5. Warrier I, Baird-Cox K, Lusher J. Use of central venous catheters in children with Haemophilia: one haemophilia treatment centre experience. Haemophilia 1997;3. 6. Blanchette VS, Al Musa A, Stain AM, Filler RM, Ingram J. Central venous access catheters in children with haemophilia. Blood Coagul Fibrinol 1996;7(Suppl. 1):S39–44. 7. Blanchette VS, Al Musa A, Stain AM, Ingram J, Fille RM. Central venous access devices in children with hemophilia: an update. Blood Coagul Fibrinol 1997;8(Suppl 1):S11–14. 8. Perkins JL, Johnson VA, Osip JM, Christie BA, Nelson SC, Moertel CL, et al. The use of implantable venous access devices (IVADs) in children with hemophilia. J Pediatr Hematol Oncol 1997;19:339–44. 9. Ljung R, van den Berg M, Petrini P, Tengborn L, Scheibel E, Kekomaki R, et al. Port-A-Cath usage in children with haemophilia: experience of 53 cases. Acta Paediatr 1998; 87:1051–4. 10. Miller K, Buchanan GR, Zappa S, Cochran C, Laufenberg J, Medeiros D, et al. Implantable venous access devices in children with hemophilia: a report of low infection rates. J Pediatr 1998;132:934–8, see comments. 11. McMahon C, Smith J, Khair K, Liesner R, Hann I, Smith O. Central venous access devices in children with congenital coagulation disorders; complications and long-term outcome. Brit J Haematol 2000;110:461–8. 12. Mumtaz H, Williams V, Hauer-Jensen M, Rowe M, HenryTillman RS, Heaton K, et al. Central venous catheter placement in patients with disorders of haemostasis. Am J Surg 2001;180:503–6. 13. Santagostino E, Gringeri A, Muca Perja M, Mannucci PM. A prospective clinical trial of implantable central venous access in children with haemophilia. Br J Haematol 1998; 102:1224–8. 14. Ingram J, Wietzman S, Greenberg ML. Complications of indwelling venous access lines in the pediatric hematology patient: a prospective comparison of external venous catheters and subcutaneous ports. J Pediatr Hematol Oncol 1991;13:130–6. 15. Wesenberg F, Flaaten H, Janssen CJ. Central venous catheter with subcutaneous port (Port-A-Cath): 8 years clinical follow-up with children. Paediatr Haematol Oncol 1993;10. €tzsch T, Carlsson M, Lethagen S. The 16. Berntorp E, Frick K, Ma Percuseal device – a new option for effective continuous prophylaxis? Haemophilia 1998;4(3):184. 17. Van Den Berg HM, Fischer K, Roosendaal G, Mauser Bunschoten EP. The use of the Port-A-Cath in children with haemophilia–a review. Haemophilia 1998;4:418–20. 18. Collins PW, Khair KS, Liesner R, Hann IM. Complications experienced with central venous catheters in children with congenital bleeding disorders. Br J Haematol 1997;99: 206–8. 19. Hothi D, Kelsall W, Baglin T, Williams D. Bacterial endocarditis in a child with haemophilia B. risks of central venous catheters. Haemophilia 2001;7. 20. Lim SH, Smith MP, Machin SJ, Goldstone AH. A prospective randomised study of prophylactic teicoplanin to prevent early Hickman catheter-related sepsis in patients receiving intensive chemotherapy for hematological malignancies. Eur J Haematol 1993;(Suppl. 54):10–3.
100
21. Ranson MR, Oppenheim BA, Jackson A, Kamthan AG, Scarffe JH. Double-blind placebo controlled study of vancomycin prophylaxis for central venous catheter insertion in cancer patients. J Hosp Infect 1990;15:95–102. 22. Medeiros D, Miller KL, Rollins NK, Buchanan GR. Contrast venography in young haemophiliacs with implantable central venous access devices. Haemophilia 1998;4: 10–5. 23. Blanchette VS, Al Trabolsi H, Stain AM, Sparling CR, Massicotte MP, Daneman A. High risk of central venous line-associated thrombosis in boys with haemophilia (abstract). Blood 1999;94:234a. 24. Journeycahe JM, Quinn CT, Miller KL, Zajac JC, Buchanan G. Catheter related deep venous thrombosis in children with haemophilia. Blood 2001;98:1717–31.
R. Ljung
25. Vidler V, Richards M, Vora A. Central venous catheterassociated thrombosis in severe haemophilia. Br J Haematol 1999;104:461–4. 26. Koerper MA, Esker S, Cobb L. Asymptomatic thrombosis of innominate vein in haemophilic children with subcutaneous venous access devices (ports) (Abstract). National Haemophilia Foundation 48th Annual Meeting, 1996. San Diego. 27. Ettingshausen C, Kurnik K, Schobess R, Kreuz W, Halimek S, Pollman H, et al. Catheter related thrombosis in children with haemophilia A. evidence of a multifactorial disease. Blood 2002;99:449–500. 28. Ljung R, Aronis-Vournas S, Kurnik-Auberger K, van den Berg M, Chambost H, Claeyssens S, et al. Treatment of children with haemophilia in Europe: a survey of 20 centres in 16 countries. Haemophilia 2000;6:619–24.