Totally implantable venous access device (TIVAD) were introduced in the late 1970’s to aid the long-term administration of intravenous therapies such as chemotherapeutic agents, antibiotics, and total parental nutrition [1]. They are used in both children and adults across a variety of diseases, however most commonly in oncology, cystic fibrosis (CF) and those requiring total parental nutrition (TPN). The number of people with TIVADs in the UK is not well documented, with most studies focusing on small cohorts in differing disease groups. The use of TIVAD’s in people with CF was first documented in 1986 [2]. The largest study to date, conducted in 2004, focused on the follow up of 452 people with CF having a TIVAD inserted [3]. This study found that complication rates in people with CF compared favorably to other populations studied, and that occlusion (21%) was in fact lower than reported in patients with malignant disease or sickle cell disease, echoed by Royle et al., study where occlusion occurred in 20% of people with CF [4].

With increased knowledge of its structure and properties, the second half of the century saw the use of unfractionated heparins in the use of prevention of venous thrombosis and central venous catheter (CVC) occlusion in renal dialysis. When CVCs, including TIVAD’s, were used for other indications, the practice of using unfractionated heparinised saline was carried across [5][6]. Heparin inhibits blood coagulation by binding to protease inhibitor antithrombin, inactivating the protease factors of the coagulation cascade. Heparin inhibits the actions of thrombin, preventing it from cleaving fibrinogen and so inhibiting the formation of fibrin. However, heparin is not a thrombolytic, so does not digest fibrin which has already formed [7]. The use of heparin does not come without risks. It is associated with the promotion of biofilms, which increases the risk of a bloodstream infection [8]-[12]. Studies have found that biofilms quickly develop once a CVC is inserted and bacteria in this biofilm can be very difficult to eradicate without line removal [8]-[13]. Due to the morbidity and mortality associated with catheter related bloodstream infections (CRBI’s) they are of significant concern for people requiring CVC’s [13][14]. This is of particular importance in recent years due to the increasing rates of CRBI’s nationally and growing resistance to the antimicrobials used to treat them [8][10][15][16]. The European Society for Clinical Nutrition and Metabolism no longer recommends heparinised saline in CVCs used for parenteral nutrition (TPN), due to the increased risk of catheter-related bloodstream infections [17].

A further associated risk is heparin induced thrombocytopenia (HIT), a severe life and limb‐threatening complication in which the immune system causes platelets to clot in the presence of heparin, resulting in low platelet levels. 1–5% of people administered heparin in any form or dose will develop HIT with thrombocytopenia [18], with unfractionated heparin having a ~10-fold greater risk of HIT than low molecular weight heparin [19][20]. Even administration at low doses, such as those used to flush CVCs, can lead to antibody formation or HIT itself [21]-[25].

In addition to this risk, there have been cases of unintentional systemic anticoagulation when using heparin flushing for CVC’s maintenance, particularly in children [26]-[30].

Furthermore, heparin, which is negatively charged, may be physically incompatible with positively charged medications, such as aminoglycoside antibiotics. If they are mixed during infusion then precipitation may occur, although the evidence base for guidelines recommending avoidance of mixing is limited [31][32].

A further consideration, as heparin is derived from porcine products (pig intestinal mucosa), its use in some religious faiths, vegans and vegetarians, requires additional consideration and informed consent must be obtained. Religious leaders only accept the use of these products if there is no alternative drug and the treatment is considered life prolonging [33][34].

As a consequence of adverse incidents, including dosing errors and HIT, in 2008, the National Patient Saftey Agency (NPSA) recommended minismising the use of heparin flush solutions in all CVCs and discontinuing its use in any peripheral intravenous catheters [35]. A Cochrane review concluded there was insufficient evidence to determine the effects of intermittent flushing of heparin versus normal saline to prevent occlusion in long term central venous catheters in infants and children as their review only found three trials, with a total of 245 participants [36]. However, pediatric guidelines state that, in the absence of convincing prospective pediatric studies, prophylaxis use of heparin for CVC occlusion prevention is not recommended in children [37][38]. A Cochrane review in 2013, which included papers from a variety of adult settings with mixed line type and duration, concluded that 0.9% saline had comparable efficacy to heparinised saline and offered a cost saving [39]. A further consensus document, collaborated by expert review of systematically collected evidence across mixed adult cohorts and settings, also concluded 0.9% saline was as appropriate as heparinised saline [40]. Despite these reports a survey of national CF centers in the UK [41] in 2017 highlighted that the majority of centers had not changed practice in line with the recommendations [35][39]. Of the centers surveyed, 90% use heparinised saline after intra-venous (IV) antibiotics administered through a totally implantable venous access device (TIVAD); 70% use heparinised saline after intravenous antibiotics administered through a peripherally inserted central catheter (PICC/midline); and 100% use heparinised saline as a 4 - 6 weekly TIVAD lock. Informal discussion with these centers revealed perseverance with historical practice rather than resistance to change [41]. The study also highlighted a reluctance to rely on the evidence as the reviews omitted papers with people with CF and did not include long term CVCs, particularly TIVADs [39]. Despite further evidence supporting the use of 0.9% sodium chloride in place of heparinised saline in mixed line type studies [42]-[44], including a further systematic review in children [45], and European CF guidelines stating there is no clear evidence to support the use of anticoagulants in TIVAD’s [46], practice within CF has remained unchanged. A TIVAD specific review in 2021 yielded limited studies [47]. Therefore, in view of possible risks associated with heparin use and ongoing research in this area, the aim of the systematic review was to assess the effectiveness of 0.9% sodium chloride versus heparinised saline to prevent TIVAD occlusion, and to discuss whether such research was applicable to the CF population.

This systematic review aimed to provide clarity on the rate of line occlusions and complications when 0.9% sodium chloride is used in place of heparinised saline in people with cystic fibrosis with TIVADs. We identified no studies in this this population, so extended the review to include all people with TIVADs. The data indicate that heparinised saline is not superior with regard to maintenance of line patency. Its use is associated with additional cost. There are insufficient data to draw conclusions regarding differences in any other adverse event associated with the use of 0.9% sodium chloride or heparinised saline. This review considers the type of flush solution used; it does not consider the optimal interval between flushes, flushing volume, or optimal technique for flushing.

Aretheseresultsapplicabletodifferentpatientpopulations? Some have theorised that advancing patient age is associated with an increase rate of thrombotic events, attributable to increases in fibrinogen and coagulation proteins [70][72]. One of the studies [56] divided the cohort to <60 and >60 and demonstrated a hazard ratio for total occlusion of 1.0 for the <60 group and 1.18 for the >60 group. Although these data are not statistically significant, if this hypothesis were correct, one would expect a reduced risk of line thrombosis in the CF cohort due to a mean lower cohort age than observed in oncology cohorts. Reduced occlusion rates seen in the Munik and Royal study may support this theory [3][4].

Howmightheparinbebeneficial? Various mechanisms of line obstruction are observed [73]: 1) catheters may stop working due to occlusion by thrombosis within the line (intraluminal occlusion); 2) thrombosis may occur within the vein, but outside the catheter, as a consequence of endothelial damage caused by the catheter; 3) a “fibrin tail” or flap may form, extending away from the tip of the catheter, which does not affect infusion through the catheter, but is drawn back and occludes the catheter during attempts at aspiration; 4) a more extensive “fibrin sheath” is formed around, and encases the tip of the line. They probably form as a reaction to endothelial injury where the tip contacts and injures the vein wall. They can develop as early as 24 hours post line insertion and increase to cover the catheter within a week [68][74]. The prevalence of fibrin sheaths increases with dwell time [75]. Heparin is an anticoagulant which inhibits the actions of thrombin, preventing the formation of fibrin. However, it is not a thrombolytic, so it will not digest fibrin which has already formed [7].

Heparin has a 60 to 90 minutes half-life in the blood, with limited data to support effectiveness once diluted in saline and in a vascular device [56][76]. Its half-life is a consequence of enzymatic rather than physical degradation, so it will persist in the lumen of a TIVAD/CVC if not exposed to blood. This persistence may be observed clinically by the prolongation of APTT clotting test when an earlier heparin lock has not been adequately flushed from a line before blood samples are taken. Paradoxically, this might mean that if exposed to blood, heparin will be digested and disappear; if no blood is present, the heparin will persist but not be needed. Therefore, the value of using a heparin lock might be greater if the line has been used previously to take blood, where some coagulation factors or platelets might remain within the lumen of an inadequately flushed line. No study considered this factor.

While it is possible that intraluminal heparin might reduce the risk of occlusion due to intraluminal thrombosis, it is less plausible it would influence the other types of occlusion due to thrombosis or fibrin outside the line. In the Munik study of 452 TIVADs [3], although occlusion type was not determined, line salvage was attempted with urokinase. This medication, would treat intraluminal thrombosis and, as a thrombolytic, be more effective than heparin against the other causes of occlusion, such as fibrin sheaths [68]. However, in this study, only 23% of blocked catheters could be salvaged by urokinase, suggesting that intraluminal thrombosis is the cause of a minority of line occlusions.

Howmighttheuseofheparinbedisadvantageous?

Statistical significance was not seen in CRBI rates in the included papers, however this has been demonstrated in wider studies [8]-[13].

Both line occlusion and infection can cause distress for patients, exposing them to additional risks from interventions to reopen/replace the line and could result in delays in treatment while venous access is determined. This can have longer-term complications and costs associated. The 2 studies [60][61] that included cost analysis both demonstrated a cost saving from stopping the use of heparin. However, both only looked at the cost of medications and omitted additional costs associated with line occlusion/infection.

The volume of heparin dose used in TIVAD care, in both oncology and CF centres, varies greatly [77][78], lacking the individualised dosing according to TIVAD catheter volume, as is recommended [46][78][79]. Although not reported in the included papers, it is important to recognise the heightened risk of increased bleeding caused by unintentional systemic anticoagulation in the CF patient group [26]-[30]. The UK National CF Registry details that 165 (2.5%) adults with cystic fibrosis reported hemoptysis at annual review in 2024 [71]. While less prevalent in the paediatric population (<5 children) [71], this risk of unintentional anticoagulation is higher due to the greater variability in size of the patient/catheter volume. Bleeding events are not only distressing for patients but also lead to a greater treatment burden, more complications, and higher associated costs [8]-[11][15].

Of the two studies which included HIT as an outcome measure (612 participants), only one case of HIT was reported [57][63]. This is significantly lower than expected with the quoted 1% - 5% for unfractionated heparin, however as HIT usually occurs within 5 to 14 days of receiving heparin, participants in these papers may have been preselected as not having HIT by virtue of previous heparin exposure [80]. A limited search yielded no data or case studies of incidents of HIT within the CF population. Additional consideration/caution should be given to those with no previous heparin exposure-for example younger patients who generally have not received heparin as prophylaxis for thromboembolic disease (due to an unfavourable risk: benefit analysis), in whom flushing a new TIVAD with heparin may indeed carry a 1% - 5% risk of HIT [18].

Alternativestoheparin? The historical inclusion of anticoagulants in line lock solutions led to consideration of the use of alternatives to heparin. Of these, citrate shows the greatest success in reducing infection rates compared to heparin [8]-[16]. There are suggestions that low dose antibiotics used alongside anticoagulants may reduce line infection rates, but the addition of another drug further increases costs and introduces extra risks, including the possibility of increased antibiotic resistance [8][10][15][16].

We believe that the comparator arm in future studies of novel lock solutions should be 0.9% sodium chloride control, rather than heparinised saline, due to the latter’s potential to increase risk of line infection and other complications. Further analysis of alternative lock solutions falls outside the scope of this review.

Thewayforward? The task of flushing TIVADs is performed almost exclusively by nursing staff. Prescription of flush solutions is performed increasingly by non-medical prescribers [81][82]. Both those administering and those prescribing a medication are accountable and have a responsibility to undertake evidence-based practice [83]-[86].

In 2012, the UK National Institute of Clinical Excellence recommended use of heparinised saline be discontinued in CVCs [35]. Non-medical prescribers tend to exercise greater caution—ensuring sound knowledge, accurate dosing, and a strong evidence base—when prescribing medications [87][88], which should result in more professionals questioning the continued use of heparinised saline in TIVADs.

This review was limited by the low number of studies in this area, particularly in pediatrics/younger cohorts. Papers were restricted to English language, which may have limited results further.

Evidence suggests that 0.9% sodium chloride is not inferior to heparin in the maintenance of TIVADs. Although there was no statistically significant difference in safety, wider research demonstrates the importance of risks associated with heparin use.

Ongoing research assessing occlusion and complications rates is of importance to monitor the long-term effect in this cohort of patients, particularly in pediatric cohorts where less data is available. The potential benefit in the subgroup of patients whose TIVAD has been used for blood sampling is worth specific consideration.

Historical practice, “what we have always done”, can be hard to change. There is a natural tendency to preserve the status quo, to require a higher level of evidence to justify a change than to continue current practice. If CVCs/TIVADs had always been flushed with 0.9% sodium chloride the evidence of potential harm, the additional cost and lack of evidence of benefit would stop clinicians from starting to use heparinised saline. With the situation reversed, we still need to ask, “Does the evidence support the continued use of heparinised saline?” While further research would strengthen this conclusion, we believe there is sufficient evidence available now, to justify a change to this historical practice.

The authors thank Dr David Waine and the adult Cystic Fibrosis Team, Plymouth Hospitals NHS Trust Hospital, Devon, UK, for their support provided throughout the review process.

Thanks to Janice Parker-Elliott, librarian at Plymouth Hospitals NHS Trust Hospital, Devon, UK, for her support in the initial search strategies of the databases.