Recommendations for developing uniform laboratory monitoring of heparinoid anticoagulants in children


  • F. NEWALL,

    1. Clinical Haematology Department, Royal Children’s Hospital, Melbourne
    2. Department of Paediatrics, University of Melbourne, Melbourne
    3. Department of Nursing, University of Melbourne, Melbourne
    4. Haematology Research Group, Murdoch Childrens Research Institute, Melbourne, Australia
    Search for more papers by this author
  • A. K. CHAN,

    1. Division of Pediatric Hematology/Oncology, McMaster University, Hamilton, Canada
    Search for more papers by this author

    1. Clinical Haematology Department, Royal Children’s Hospital, Melbourne
    2. Department of Paediatrics, University of Melbourne, Melbourne
    3. Haematology Research Group, Murdoch Childrens Research Institute, Melbourne, Australia
    Search for more papers by this author

    1. Clinical Haematology Department, Royal Children’s Hospital, Melbourne
    2. Department of Paediatrics, University of Melbourne, Melbourne
    3. Haematology Research Group, Murdoch Childrens Research Institute, Melbourne, Australia
    Search for more papers by this author

Fiona Newall, Clinical Haematology Department, Royal Children’s Hospital, Flemington Rd, Parkville, Victoria 3502, Australia.
Tel.: +61 3 9345 5762; fax: +61 3 9349 1819.

Introduction and rationale

Pediatric-specific pharmacokinetic (PK) models for unfractionated heparin (UFH) and low molecular weight heparins (LMWHs) have only recently been established [1–4]. As in adults, the optimal PK profiles for both UFH and LMWHs are based on weight; however, the parameter estimates generated within pediatric populations differ significantly from those reported for adult populations. For example, in neonates, the clearance parameter estimate for UFH is 1.37 ± 0.46 mL kg−1 min–1 [5], whereas the equivalent measure in adults is 0.43 ± 0.09 mL kg−1 min–1 [5]. Not surprisingly, the clearance parameter estimate for UFH in children is midway between those for neonates and adults, at 0.6 ± 0.4 mL kg−1 min–1 [1]. Similar age-related parameter estimate differences are observed for the volume of distribution and half-life of UFH in neonates, children, and adults [1,5]. With respect to LMWHs, the contribution of limited subcutaneous tissue to drug absorption may impact on time to peak levels and half-life across the spectrum of pediatric ages; however, the extent of this contribution requires further investigation.

This variability in age-related PK parameter estimates supports the premise that the pharmacodynamic profile for anticoagulants in children will be different from that in adults. This has certainly been demonstrated for LMWHs and UFH. Recent evidence has demonstrated that significant age-related differences in LMWH response exist across childhood, as measured by the anti-factor Xa assay [6]. In exploring the age-related differences in LMWH response across childhood, Ignjatovic et al. [7] demonstrated that, although in vivo differences clearly exist, spiking plasma from children of different ages with LMWH does not result in similar age-related differences. This disparity between in vivo and in vitro findings supports the need to obtain pharmacodynamic evidence from true clinical studies rather than laboratory-based in vitro experiments. This difference may be attributable to the interplay of several factors, including the medical complexity of the patients impacting on the hemostatic system, response to the medicine, and variable competitive plasma binding. Similar evidence exists with respect to age-dependent differences in the response to UFH. In vivo studies have demonstrated age-related differences in the dose–response relationship, a variable ratio of anti-FIIa to anti-FXa activity, and age-dependent differences in the impact of UFH on tissue factor pathway inhibitor release [8–13]. Furthermore, the contribution of renal and hepatic dysfunction to the variable clinical response to LMWH and UFH requires consideration.

Substantial evidence therefore supports the use of pediatric-specific PK and laboratory-based pharmacodynamic profiles of these commonly used anticoagulants in children. Nonetheless, clinicians still lack robust pharmacodynamic models for all of the commonly used anticoagulant agents that incorporate laboratory and clinical outcome measures. As a result, the current recommendations regarding anticoagulant dosing or laboratory monitoring in children are not based on appropriately robust levels of evidence. Rather, clinical practice guidelines for the management of anticoagulant therapies in children continue to be extrapolated from adult evidence, with occasional pockets of pediatric-specific evidence.

Recommendations and discussion

This position paper recommends that a strategic and collaborative approach should be undertaken to establish robust pharmacodynamic models for UFH and LMWH, as the most commonly used anticoagulant agents in pediatric practice [14]. These models must ultimately incorporate concomitant laboratory and clinical outcome data. A stepwise approach to the generation of this evidence should include the following:

  • 1 Justification for monitoring of anticoagulant effect.
    • Clinical trials of heparin-based anticoagulant management in adults have not demonstrated superiority of clinical outcomes for monitored therapy relative to unmonitored therapy. Improved clinical outcomes have been demonstrated through the introduction of weight-adjusted dosing vs. non-weight adjusted dosing [15–17]. Given the inherent challenges and inconsistencies associated with all monitoring assays used in the management of heparinoid therapies, one cannot rule out the possibility that weight-adjusted dosing without laboratory monitoring may achieve similar safety and efficacy outcomes as monitored therapy.

  • 2 Identification of the optimal assay method.
    • Activated partial thromboplastin time and anti-FXa, the two most commonly used measures of UFH effect in children, have a poor correlation with UFH concentration [11,18]. Additionally, the use of anti-FXa assay reagents with exogenous hemostatic proteins (antithrombin and dextran sulfate) has been shown to produce significant variations in the measured UFH effect [19]. Several studies have suggested that the thrombin clotting time may offer improved specificity in UFH monitoring; however, further evidence is required to confirm this [10,20,21]. With respect to LMWH monitoring, the previously mentioned variation in measured anti-FXa activity between different reagents when UFH is monitored is also likely to be observed when LMWHs are monitored, although this has yet to be confirmed [22]. The use of reagents that introduce exogenous hemostatic proteins into the assay process are not ideal as they are not physiologically relevant. Clinical outcome studies must be developed to demonstrate the clinical utility of any anticoagulant monitoring assay in children.

  • 3 Determination of indication-specific and medication-specific therapeutic ranges.
    • Once the optimal assay for anticoagulant monitoring has been determined, clinical trials evaluating appropriate management strategies for anticoagulant therapies in children can be initiated. Given the challenges of conducting robust clinical trials of anticoagulants in children [23], international, highly coordinated and adequately funded studies need to be developed. Only through the development of such trials will evidence supporting the optimal dosing and monitoring strategies for anticoagulant therapies in pediatric practice be generated.

Additionally, all novel anticoagulant agents commencing licensing processes for use in pediatrics should similarly require the development of robust, pediatric-specific PK and pharmacodynamic modeling.

In conclusion, the assumption that therapeutic ranges can simply be extrapolated from adult evidence and applied to pediatric practice does not hold true. A systematic approach is required to justify the use of laboratory monitoring of anticoagulation in children.

Disclosure of conflict of interests

The authors state that they have no conflict of interest.