Late rebound of dabigatran levels after idarucizumab reversal in two patients with severe renal failure

Editor, An 87-year-old woman with a history of heart failure, aortic stenosis, chronic kidney disease [creatinine clearance (CrCl), usually in the region of 56 ml min−1) and atrial fibrillation, for which she received dabigatran at 110 mg twice daily, presented to the emergency department with a traumatic femoral fracture. She had taken her last dose of dabigatran 17 h before admission. On arrival, severe renal impairment (CrCl, 25 ml min−1) and anaemia with a haemoglobin concentration of 101 g l−1 were observed. Her prothrombin time (PT) was 0.57 with an INR of 1.35 (Thromborel S, Siemens, Marburg, Germany; manufacturer's reference values for INR are 0.70 to 1.2). No other coagulation tests were performed at this time. Six hours after admission, she developed severe haemorrhagic shock (haemoglobin level, 57 g l−1). Because of this life-threatening condition, she was given three units of red blood cells and 1000 units of prothrombin complex concentrate (PCC), and 1 hour later, she was treated with 5 g of idarucizumab. On day 1 postinjury (12 h after idarucizumab administration), thrombin time (TT; BC Thrombin, Siemens, Marburg, Germany) was 24.3 s (normal range, 14 to 21 s) and activated partial thromboplastin time (aPTT; Pathromtin SL, Siemens, Marburg, Germany) was 45.1 s (normal range 26 to 36 s; Fig. 1). Plasma dabigatran concentration was below the lower limit of quantification (8 ng ml−1), as assessed by the ‘in-house’ diluted thrombin time (dTT; Test Thrombin, Siemens, Marburg, Germany; lower limits of detection and quantification were 3 ng ml−1 and 8 ng ml−1, respectively).1 Analyses of PT, aPTT and TT were performed using BCS XP analyser (Siemens, Marburg, Germany), whereas dTT was determined using CS2100i analyser (Sysmex, Kobe, Japan). Internal fixation of the femur was performed under general anaesthesia without major blood loss. Twenty-four hours after surgery (37 h after idarucizumab administration), the patient started to bleed again (haemoglobin level, 84 g l−1; TT, 23.4 s; aPTT, 58.5 s; dTT, < 8 ng ml−1; Fig. 1). On day 3 postinjury, the patient deteriorated because of cardiac failure, accompanied by acute renal failure, as manifested by oliguria and a CrCl less than 15 ml min−1. Because of hypervolaemia, she underwent 3 h of haemodialysis using a high-flux membrane with blood flow and dialysate flow rates of 250 and 500 ml h−1, respectively and regional citrate anticoagulation [Isosmotic trisodium citrate solution (102 mmol l−1) was infused into the arterial side of the dialyzer and of 5% CaCl was infused into the venous return line at a rate of 0.5 ml min−1]. Thus, no systemic anticoagulant was used. One hour later (74 h after idarucizumab administration), the anticoagulant activity of dabigatran was evident again (dTT, 130 ng ml−1), which was mirrored by a concomitant increase in TT (>150 s) and aPTT (107.5 s). The patient was intubated and referred to the surgical ICU. Because of the clinical and laboratory evidence of bleeding, elevated dabigatran levels and anticipated delayed clearance because of renal failure, additional haemodialysis was performed for dabigatran removal on days 4 and 5 postinjury. The renal function improved and dabigatran levels gradually declined 6 h after each haemodialysis. On day 8 postinjury, dabigatran levels were below the lower limit of quantification (Fig. 1). Because of the high thromboembolic risk, unfractionated heparin was initiated and the dose adjusted to an aPTT value of 50 to 55 s. No recurrent bleeding was observed. The patient was discharged from the ICU on day 45 postinjury. Informed consent for publication was obtained from the patient's guardian.Fig. 1: Activated partial thromboplastin time, thrombin time, diluted thrombin time, creatinine clearance and the time of idarucizumab administration in case 1. aPTT, activated partial thromboplastin time; CrCl, creatinine clearance; D, haemodialysis, dTT, diluted thrombin time; OP, surgical fixation of a femoral fracture; s, seconds; TT, thrombin time.A 74-year-old woman with arterial hypertension, peripheral arterial disease and atrial fibrillation, for which she received dabigatran 110 mg twice daily, was admitted to the hospital with critical right leg ischaemia, a gangrenous third toe and sepsis. She had taken her last dose of dabigatran 12 h before admission. Although, her CrCl had been 80 ml min−1 at home before the acute episode, on admission, severe renal failure was observed with a CrCl of 27.4 ml min−1. Her dTT was 217 ng ml−1 (36 h after the last dose of dabigatran). She was given 5 g of idarucizumab because transmetatarsal amputation of the third toe was planned. Eighteen hours after idarucizumab administration, dTT was below the limit of quantification and surgery was performed uneventfully. The following day, 42 h after idarucizumab administration, dTT increased to 62 ng ml−1 (i.e. 28% of the preidarucizumab level; Fig. 2). Percutaneous transluminal angioplasty with stenting of the right femoral artery was performed, and dual antiplatelet therapy was initiated. After the procedure, a large haematoma occurred at the puncture site with a concurrent fall in haemoglobin concentration to 20 g l−1. After intravenous hydration, the renal function gradually improved. On day 6 after admission, dTT was 13 ng ml−1. On discharge from hospital (day 24), her CrCl was 44 ml min−1 and she was maintained on dual antiplatelet therapy. Informed consent for publication was obtained from the patient.Fig. 2: Diluted thrombin time, creatinine clearance and time of idarucizumab administration in the second patient. CrCl, creatinine clearance; dTT, diluted thrombin time; OP, surgical fixation of a femoral fracture.Dabigatran is a direct thrombin inhibitor with an effective anticoagulant activity and is used in patients with nonvalvular atrial fibrillation.2 Dabigatran and its metabolites are mainly eliminated by the kidneys. In patients with renal impairment, there is a significant increase in the elimination half-life of dabigatran, which results in accumulation of the drug and excessive anticoagulation.3 In this report, we present two patients with normal baseline CrCl more than 50 ml min−1 who were treated with dabigatran (110 mg twice daily) but who then developed acute renal failure in association with acute illness/trauma. Because of the expected delay in dabigatran clearance on account of the renal failure, and the life-threatening bleeding in the trauma patient and the imminent surgery in the septic patient, both patients received 5 g of idarucizumab, in accordance with the International Society on Thrombosis and Haemostasis recommendations.4 Idarucizumab binds dabigatran with an affinity that is 350 times higher than that observed with thrombin.5 In addition, the trauma patient initially received red blood cells and PCC because of massive blood loss. Blood products, fresh frozen plasma and PCC were also coadministered in 25% of the patients in the REVERSE-AD study.5 At present, it is unclear whether idarucizumab should be used as monotherapy, particularly in patients with dilutional or trauma-induced coagulopathy.6 Idarucizumab reverses the anticoagulant effects of dabigatran promptly and completely in the majority of dabigatran-treated patients with life-threatening bleeding or who require emergency invasive procedures.5 Pollack et al.5 showed that there was a subsequent increase in dabigatran concentrations and clotting times in a minority of the patients at 12 and 24 h after idarucizumab administration; however, further monitoring was not performed. We observed complete normalisation of coagulation tests in both patients after idarucizumab administration but on the 2nd or 3rd day after idarucizumab administration, dabigatran levels increased in both patients to levels typically observed in patients on dabigatran treatment.7 At these times, the renal function of both patients was still severely impaired. Redistribution of dabigatran from the extravascular to the intravascular compartment can contribute to this rebound anticoagulant effect. The potential for heparin contamination of the dTT assay during blood sampling and haemodialysis was excluded. Glund et al.8 reported an increased exposure and decreased clearance of idarucizumab in otherwise healthy volunteers with mild and moderate renal impairment. However, our patients were critically ill and had severe progressive renal impairment. Therefore, and notwithstanding the results of Glund et al.,8 it is likely that the total body dabigatran load exceeded the capacity of a single dose of idarucizumab to achieve complete and sustained neutralisation of dabigatran. Other studies have reported the reappearance of dabigatran activity after idarucizumab reversal; however, in these patients, dabigatran concentration on admission was extremely high and a rebound effect of dabigatran appeared within 24 h.9,10 Unfortunately, in the trauma patient, we did not assess dTT on admission to the emergency department before idarucizumab administration. PT is relatively insensitive to plasma dabigatran level and aPTT has a low specificity for detection of residual dabigatran activity in the presence of dilutional and trauma-induced coagulopathy. Therefore, they are not recommended. In conclusion, we suggest assessment of dabigatran activity in bleeding patients on admission and, in addition, in patients with severe renal failure, serial measurements of dabigatran concentration during the first week after idarucizumab administration. A specific test, such as dTT, is preferred. If dTT is unavailable, TT can provide a qualitative estimation of the effects of dabigatran. Confirmation of adequate and sustained dabigatran reversal is essential for the management of bleeding, planning for urgent invasive procedures, decision making for additional doses of idarucizumab, haemodialysis and the timing of resuming antithrombotic treatment in patients with an underlying predisposition to thrombosis. Acknowledgements relating to this article Assistance with the letter: none. Financial support and sponsorship: none. Conflicts of interest: none.