Chronic kidney disease (CKD), a reduction in renal function with decreasing glomerular filtration rate (GFR) is increasing in incidence as well as in prevalence due to a growing elderly population. The progressive reduction in GFR to less than 60ml/min (modest renal insufficiency; mRI) or severe renal insufficiency (sRI; with GFR <30ml/min) may lead to end stage renal disease (ESRD) with a requirement for chronic haemodialysis (CHD).
In parallel with progressing age, there is an increase in CKD, elective and non-elective surgeries, non-valvular atrial fibrillation (AF) and venous thromboembolism (VTE). Of note, CKD and AF as well as CKD and VTE are not independent conditions, but, in both situations, positive bidirectional relationships have been proved.1,2
Patients undergoing major surgery have an elevated risk for thromboembolic events. In these patients and those suffering any VTE, anticoagulant therapy is highly effective in the primary and secondary prevention of thromboembolism. Yet, this efficacy of anticoagulant drugs also exposes patients to an increased risk of anticoagulant-associated bleeding.3
This short review will neither address the appropriate anticoagulation in the context of CHD nor the problem of how to adapt anticoagulation in acute kidney failure, but will discuss specific aspects of anticoagulant treatment in patients with CKD at risk for acute or recurrent VTE.
Anticoagulant drugs and renal insufficiency
For a long time parenteral drugs such as unfractionated heparin (UFH), low-molecular-weight heparins (LMWH), or fondaparinux (FPX) were used for acute initiation of anticoagulation therapy, which was then switched to an oral vitamin K antagonist (VKA) such as warfarin. Nowadays direct oral anticoagulants (DOACs) have become available for primary and secondary prevention as recently reviewed by Moss and colleagues.4
These anticoagulant drugs do not only differ in their mode of action and the time course of anticoagulant activity but also in the degree of drug–drug interactions as well as in the dependency of their pharmacokinetic parameters on kidney function (Table 1). Most anticoagulant drugs (LMWH, FPX, DOACs) can accumulate in patients with impaired renal capacity.5
The recent randomised Phase III landmark studies in VTE6 and the supporting evidence in AF,7 demonstrated the efficacy and safety of anticoagulant drugs but excluded patients with sRI or even ESRD. With the exception of LMWH – where some data are available8,9 – recommendations on how to treat patients with markedly reduced GFR having an urgent need for anticoagulation are based mainly on considerations of pharmacokinetic data, extrapolation of clinical study results in subgroup analyses of patients with mRI, and case series.
Patients with CKD have an elevated risk for bleeding with declining GFR, even in the absence of antithrombotic drugs,10 but also seem to have an elevated risk of thromboembolism.2 Therefore, a thorough evaluation of the patient’s individual risk profile concerning thromboembolic events and bleeding complications has to be recommended prior to the choice of drug and dose for anticoagulation. Moreover, this risk profile should be re-assessed on a regular basis.
Due to different pharmacodynamic properties of anticoagulant drugs (Table 1), the risk of accumulation, resulting in an increased bleeding risk, differs among the available drugs and needs to be considered in patients with an estimated GFR of <50ml/min (Table 2).
Table 1 (Click on image to enlarge)
Table 2 (Click on image to enlarge)
The miscellaneous LMWH have different molecular weight distributions, affecting their excretion by the kidneys, and resulting in an inverse relationship between molecular weight and renal clearance.11 Moreover, FPX has a longer half-life and a higher risk of accumulation in patients with CKD compared with LMWH.
For DOACs, the thrombin inhibitor dabigatran has the highest risk of accumulation in patients with impaired GFR, whereas the oral factor Xa inhibitors have a lower risk for accumulation.7 For UFH and VKA, the risk of accumulation is very low, but therapeutic anticoagulation demands for laboratory controls – via aPTT or anti-factor Xa activity for UFH and via INR for VKA – thus minimising the risk of unintended over-anticoagulation. Monitoring of the anticoagulant activity of LMWH, FPX or DOACs are not regularly recommended but available.
Life-threatening complications such as heparin-induced thrombocytopenia type II and warfarin-induced purpura fulminans do rarely occur and their incidences do not seem to be influenced relevantly by CKD. Warfarin-related nephropathy is another seldom complication, which may stimulate the progression of CKD, especially when the INR is above therapeutic range. Interestingly, a dabigatran-related nephropathy has recently been reported.12 Yet, reports also indicate calciphylaxis related to the use of warfarin in patients with CKD.13 Thus, VKA in patients suffering from CKD is controversially discussed among nephrologists, unless there is an absolute indication, such as after artificial heart valve replacement.13
Primary prevention of VTE in CKD patients undergoing surgery
Perioperative morbidity and mortality are common issues in the management of patients with kidney diseases undergoing elective or emergency surgery. Many of these patients are to be initiated on postoperative anticoagulation for VTE prevention. Pharmacological thromboprophylaxis reduces the risk of VTE, most convincingly demonstrated after elective total hip or knee replacement surgery.
For patients undergoing these procedures, randomised clinical trials and meta-analysis of DOACs demonstrate an overall comparable efficacy and safety for DOACs compared to LMWH.14 As long as GFR is >30ml/min, standard dosages of LMWH or most DOACs can safely be applied postoperatively for 10 to 35 days. Concerning FPX, a reduced dosage of 1.5mg once daily is recommended for patients with a limited GFR between 50 and 20ml/min. For patients with sRI standard doses of LMWH may be safe, but convincing prospective study data are missing.8 Dose reductions by 30–50% and/or monitoring of the anti-factor-Xa activity are suggested in patients with GFR between 30 and 15ml/min.
The subcutaneous application of 3 x 5000IU or 2 x 7500IU UFH per day is a traditional and effective procedure in patients with sRI or ERD.
Outside the setting of elective joint arthroplasty, meaningful subgroup analyses of patients with CKD are rare. Therefore, recommendations derived from joint replacement surgeries are considered references for postoperative VTE prevention also in other indications.
Many of those – elderly – patients with CKD undergoing surgery are on long-term anticoagulation therapy prior to the operation due to AF or VTE. Nowadays, most of these patients receive DOACs. Yet, there still is a relevant proportion of patients treated with VKA for these indications or others. Although minor procedures can be performed in most patients while on trough levels of DOACs or in the lower part of the therapeutic INR-range, major surgeries or interventions with an increased risk of haemorrhage require normal hemostasis. For DOACs an interruption for 2–5 days (Table 2) depending on the specific DOAC as well as on renal function seems to be a safe strategy. Bridging of VKA anticoagulation by LMWH in the context of AF has recently become a matter of discussion.15 Yet, it needs to be stressed that in patients with sRI or ERD, having a clear cut indication for anticoagulation therapy, bridging via application of UFH may be less convenient, but safer than the use of LMWH. Postoperative anticoagulation in these patients may be initiated with a prophylactic dose for some days before – stepwise – escalation to the therapeutic dose, depending on the postoperative bleeding risk.
Full dose anticoagulation therapy in patients with CKD and VTE
Prospective randomised trials in general populations with VTE have demonstrated that anticoagulation therapy significantly decreases recurrent VTE to a larger extend than it increases hemorrhagic complications.6 This beneficial effect of treatment has been confirmed in observational studies and registries. In large registration studies patients with acute VTE (or AF) and mRI have been included and subgroup analysis demonstrated a comparable ‘net-benefit’ of therapeutic anticoagulation with numerically higher rates of thromboembolic events and bleeding complications compared with patients with a GFR >60ml/min.5-7 However, patients with sRI or CHD were largely excluded.
Thus, prospective evidence as well as systematic reviews and meta-analyses for patients suffering from sRI or ESRD/CHD and acute VTE are not available. However, tens of thousands of patients with AF and CHD treated with warfarin have been analysed with heterogeneous results.6,17 Thus, any clear benefit of VKA in the CHD population seems to be questionable. The lack of obvious benefits supports the need for randomised trials in this setting as well as in patients with sRI, ESRD/CHD and acute VTE.
In this context, it should be noted that embolism in AF is a permanent risk, whereas the risk of recurrent VTE decreases with time from the initial event. Thus, a transfer of study results in patients with AF to those with VTE needs critical reflection.
A retrospective analysis following 1509 patients with VTE and sRI for three years demonstrated higher risks for relapse, major bleedings and mortality.18 The available data comparing the initial anticoagulation in sRI or ESRD patients with acute VTE did not demonstrate any benefit for UFH compared with LMWH with regard to efficacy or safety but hint at an increased mortality in patients with sRI treated with UFH compared with LMWH.19 Nevertheless, prospective randomised clinical trials are still unavailable. Until direct evidence is available, highly individual clinical decision-making must be recommended, especially with regard to the duration of full-dose anticoagulation therapy, because treatment efficacy in most patients suffering from acute VTE can be monitored by clinical, technical (ultrasound) and laboratory (D-dimer) investigations. Thus, a tailored duration and intensity of anticoagulation can be determined, thereby limiting the risk of haemorrhage due to prolonged anticoagulation.
The availability of DOACs provides an alternative oral anticoagulation solution.6 Again, subjects with sRI were excluded in the pitoval trials in VTE (and AF) with the thrombin inhibitor dabigatran (GFR <30ml/min) and the factor Xa inhibitors apixaban (GFR <25ml/min), edoxaban (GFR <30ml/min) or rivaroxaban (GFR <30ml/min).4-6
Based on published investigations of pharmacokinetics in patients with CHD, dose recommendations have been issued for ESRD.20 For apixaban, the regular 5mg twice daily dose (2.5mg twice daily in patients > 80 years or < 60kg body weight) is suggested. Concerning rivaroxaban, a daily dose of 10mg in CHD patients resulted in drug exposure parameters similar to the findings for 20mg once a day in healthy volunteers. For edoxaban, a 15mg dose once daily in ESKD patients was found to be equivalent to 60mg in mRI. None of the factor Xa inhibitors were eliminated to a relevant degree by dialysis. Overall there may be a benefit for oral anticoagulation with anti-factor Xa DOACs compared with warfarin in patients with sRI or ESRD. Thus the door has been opened to a cautious application of these DOACs in patients with severe CKD. While data from prospective trials are lacking, we recommend monitoring of the anticoagulant effect by measuring drug-calibrated anti-factor Xa activity – by determination of trough concentrations, for the first time after three doses – if DOACs are used in patients with GFR <30ml/min.
CKD continues to be a major contributory factor for morbidity and mortality in the context of VTE. Anticoagulation is effective in preventing and treating thromboembolisms. The available body of evidence convincingly demonstrates that treatment with parenteral or oral anticoagulant drugs is not only effective and safe in patients with normal renal function, but also in those with slightly or moderately reduced renal function. In patients with sRI or ESRD, anticoagulant treatment for acute VTEs is challenging because prospective randomised trials, evaluating efficacy and safety in these more advanced stages of renal insufficiency are lacking. Some evidence from retrospective analyses and small prospective Phase II studies support the use of anticoagulant drugs in patients with acute VTE and sRI. In patients undergoing CHD, the benefits and harms of prolonged or long-term anticoagulation therapy are less clearly balanced.
Whereas a short-term therapeutic anticoagulation therapy for an acute VTE is recommended, the decision to proceed to a prolonged therapeutic anticoagulation needs to be critically evaluated individually in the context of the thromboembolic and haemorrhagic risks. Despite the long-standing clinical practice of using UFH and VKA in patients with ESRD, available but limited data are in favor of the use of LMWH instead of UFH. When long-term anticoagulation is indicated for VTE anti-factor Xa DOAKs may be the better choice as compared to VKA. There is a high clinical need for prospective randomised clinical trials evaluating anticoagulation strategies in patients with sRI and ESRD.
1 Bansal N et al. Incident atrial fibrillation and risk of end-stage renal disease in adults with chronic kidney disease. Circulation 2013;127(5): 569–74.
2 Ocak G et al. Role of hemostatic factors on the risk of venous thrombosis in people with impaired kidney function. Circulation 2014;129(6):683–91.
3 Sardar P et al. Novel oral anticoagulants in patients with renal insufficiency: a meta-analysis of randomized trials. Can J Cardiol 2014;30(8):888–97.
4 Moss AS, Dimitropoulous G, Lip GYH. Clinical implications, benefits and pitfalls of using and reversing non-vitamin K antagonist oral anticoagulants. Expert Rev Hematol 2017;10(9): 833–45.
5 Harenberg J et al. Anticoagulation in patients with impaired renal function and with haemodialysis. Anticoagulant effects, efficacy, safety, therapeutic options. Hamostaseologie 2015;35(1):77–83.
6 Gomez-Outes A et al. Direct oral anticoagulants in the treatment of acute venous thromboembolism: a systematic review and meta-analysis. Thromb Res 2014;134(4):774–82.
7 Nielsen PB et al. Renal function and non-vitamin K oral anticoagulants in comparison with warfarin on safety and efficacy outcomes in atrial fibrillation patients: a systemic review and meta-regression analysis. Clin Res Cardiol 2015;104(5):418–29.
8 Lim, W et al. Meta-analysis: low-molecular-weight heparin and bleeding in patients with severe renal insufficiency. Ann Intern Med 2006;144(9):673–84.
9 Hoffmann P, Keller P. Increased major bleeding risk in patients with kidney dysfunction receiving enoxaparin: a meta-analysis. Eur J Clin Pharmacol 2012;68(5):757–65.
10 Ocak G et al. Chronic kidney disease and bleeding risk in patients at high cardiovascular risk: a cohort study. J Thromb Haemost 2018;16(1):65–73.
11 Gouin-Thibault I et al. Safety profile of different low-molecular weight heparins used at therapeutic dose. Drug Saf 2005; 28(4):333–49.
12 Narasimha Krishna V et al. Oral anticoagulants and risk of nephropathy. Drug Saf 2015; 38(6):527–33.
13 Hasegawa H. Clinical assessment of warfarin therapy in patients with maintenance dialysis-clinical efficacy, risks and development of calciphylaxis. Ann Vasc Dis 2017;10(3).
14 Adam SS et al. Comparative effectiveness of new oral anticoagulants and standard thromboprophylaxis in patients having total hip or knee replacement: a systematic review. Ann Intern Med 2013; 159(4):275–84.
15 Douketis JD et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med 2015;373(9):823–33.
16 Harel Z et al. Warfarin and the risk of stroke and bleeding in patients with atrial fibrillation receiving dialysis: A systematic review and meta-analysis. Can J Cardiol 2017;33(6):737–46.
17 Van Der Meersch H, De Bacquer D, De Vriese AS. Vitamin K antagonists for stroke prevention in hemodialysis patients with atrial fibrillation: A systematic review and meta-analysis. Am Heart J 2017;184:37–46.
18 Parikh AM et al. Venous thromboembolism in patients with reduced estimated GFR: a population-based perspective. Am J Kidney Dis 2011;58(5):746–55.
19 Trujillo-Santos J et al. Low-molecular-weight or unfractionated heparin in venous thromboembolism: the influence of renal function. Am J Med 2013;126(5):425–434 e1.
20 Chan KE et al. Nonvitamin K anticoagulant agents in patients with advanced chronic kidney disease or on dialysis with AF. J Am Coll Cardiol 2016;67(24): 2888–99.