- Email: [email protected]
Recurring Extracorporeal Circuit Clotting During Continuous Renal Replacement Therapy in Fungal Sepsis: Successful Treatment With Argatroban
CASE REPORT
Recurring Extracorporeal Circuit Clotting During Continuous Renal Replacement Therapy in Fungal Sepsis: Successful Treatment With Argatroban Lee M. Ferguson, MD, Albert W. Dreisbach, MD, Éva Csongrádi, MD, Luis A. Juncos, MD and Tibor Fulop, MD
Abstract: The relative effectiveness of anticoagulation strategies during continuous renal replacement therapy (CRRT) may vary according to the clinical circumstances. In this study, the case of a 46-year-old man who developed fungal mediastinitis with the pathogen Scedosporium prolificans after coronary bypass surgery is reported. Numerous debridements and multiple antifungal agents were not effective in this patient. Miltefosine, a non-Food and Drug Administration-approved agent, was started after institutional review board request and approval. CRRT was initiated with regional citrate anticoagulation (RCA) for clinical sepsis with acute kidney injury. Subsequently, crescendo clotting of the extracorporeal circuit (ECC) occurred. Multiple interventions, including escalating RCA, adding increasing heparin to RCA and exchanging the dialysis catheter, were not effective. Argatroban anticoagulation was started without further ECC clotting, and the patient recovered from both acute kidney injury and septic shock, despite continued miltefosine administration. Sepsis may contribute to recurrent ECC clotting. Argatroban, a direct thrombin inhibitor, had a disproportionate effectiveness to maintain ECC patency in this patient. Key Indexing Terms: Argatroban; Continuous renal replacement therapy; Miltefosine; Regional citrate anticoagulation; Sepsis. [Am J Med Sci 2013;345(3):256–258.]
I
INTRODUCTION
n critically-ill patients, extracorporeal circuit (ECC) clotting is a frequent complication of continuous renal replacement therapy (CRRT). Traditionally, this is prevented by using regional citrate anticoagulation (RCA) or prefilter unfractionated heparin.1,2 Argatroban is a direct thrombin inhibitor whose role in CRRT anticoagulation has been limited to patients with either heparin-induced thrombocytopenia or antithrombin III (AT3) deficiency.3–6 Here, we report a case of recurrent ECC clotting in a patient with fungal sepsis, despite RCA and the addition of unfractionated heparin. After switching to argatroban as the primary anticoagulant, the patient was able to undergo CRRT without further ECC clotting.
CASE REPORT Our patient is a 46-year-old Caucasian man with type II diabetes mellitus, hypertension and coronary artery disease who underwent a 4-vessel coronary artery bypass graft after having a myocardial infarction. Postoperatively, the patient developed From the Department of Medicine (LMF) and Division of Nephrology, Department of Medicine (AWD, LAJ, TF), University of Mississippi Medical Center, Jackson, Mississippi; First Department of Medicine (ÉC), Medical and Health Science Center, University of Debrecen, Debrecen, Hungary; and Department of Physiology and Biophysics (LAJ), University of Mississippi Medical Center, Jackson, Mississippi. Submitted June 14, 2012; accepted in revised form August 27, 2012. The authors have no financial or other conflicts of interest to disclose. Correspondence: Lee M. Ferguson, MD, Department of Medicine, University of Mississippi Medical Center, 2500 North State Street, L 504, Jackson, MS 39216-4505 (E-mail: [email protected]).
256
sternal osteomyelitis, with sternal cultures growing an unusual fungal pathogen, Scedosporium prolificans. Over the next year, the patient continued to have recurring fungal sternal osteomyelitis, despite multiple courses of various intravenous antifungal medications and numerous sternal debridements (48 in all). He was subsequently left with a complete sternotomy with definitive closure by omental flap and skin grafting. Despite all these measures, he experienced yet another recurrence of fungal osteomyelitis and mediastinitis. Therefore, after obtaining institutional review board approval, the patient was placed on the experimental antifungal drug, miltefosine (MILT). Two days after initiation of MILT, he developed hypotension, clinical sepsis and acute kidney injury (AKI), necessitating temporary hemodialysis (HD). MILT was thought to have contributed to his AKI and was discontinued during the admission. He was subsequently diagnosed with gangrenous cholecystitis, necessitating emergency surgical cholecystectomy; he then gradually recovered. Two months later, the patient was readmitted with purulent drainage from his sternal graft site. The initial physical examination revealed a temperature of 98.6⁰F, a heart rate of 72 beats per minute, a blood pressure of 96/54 mm of Hg, a respiratory rate of 18 breaths per minute and an O2 saturation of 97%. The sternal graft site had erythema around the edges and edema of the flap. Bilateral lower extremities had moderate pitting edema. The initial laboratory values were as follows: white blood cell count, 5,900 per cubic millimeter; hematocrit, 22.9%; platelet count, 170,000 per cubic millimeter; glucose level, 59 mg/dL; sodium level, 135 mmol/L; potassium level, 4.4 mmol/L; chloride level, 99 mmol/L; bicarbonate level, 29 mmol/L; blood urea nitrogen level, 46 mg/dL; and creatinine level, 2.24 mg/dL. The patient was placed on voriconazole and caspofungin for presumed recurrent fungal mediastinitis and daptomycin for possible methicillin-resistant Staphylococcus aureus infection. Additionally, oral MILT was reintroduced to his regimen. Over the next 2 days, the patient became oliguric and developed volume overload with worsening pulmonary edema. Given that the patient had a rising serum creatinine, he was again diagnosed with AKI complicating his chronic kidney disease. A bilateral renal ultrasound revealed increased cortical echogenicity of both kidneys and no evidence of hydronephrosis. Despite aggressive diuretic therapy, the patient continued to have worsening volume overload and was transferred to the medical intensive care unit for the initiation of CRRT. At the time of transfer, serum creatinine level measured 4.4 mg/dL; the blood urea nitrogen level was 63 mg/dL and platelets were 269,000 per cubic millimeter. A temporary right internal jugular HD catheter was placed, and CRRT was started with RCA. CRRT used an acid citrate dextrose solution A at a prefilter flow rate of 300 mL/ hr, a blood flow rate of 200 mL/min, a hemofiltration rate of 2500 mL/hr with premixed calcium-free replacement fluid
The American Journal of the Medical Sciences
Volume 345, Number 3, March 2013
Argatroban During CRRT
(split 50%–50% between prefilter and postfilter dilution), a systemic calcium chloride infusion, and an effective filtration fraction of 11% during the procedure. Initiation of CRRT was complicated by hypotension, and the patient was subsequently given a 700-mL bolus of intravenous normal saline and started on an intravenous norepinephrine infusion for septic shock. CRRT was started and lasted for approximately 1 hour before ECC clotting occurred. Blood clots were found in CRRT tubing and HD catheter ports. Despite escalating RCA and adding prefilter heparin, the patient continued to have recurrent ECC clotting. Table 1 shows the ECC clotting times with the various anticoagulation methods. Fibrinogen levels were only mildly elevated at 401 and 485 mg/dL (reference range, 180–400 mg/dL). Serum procalcitonin level was markedly elevated at 15.81 ng/mL (reference range, 0.00–0.05 ng/mL). After multiple failed restarts of CRRT secondary to ECC clotting, the patient was placed on a critical illnessadjusted dose of argatroban of 0.5 mg/kg/min. Activated partial thromboplastin times were checked, initially every 2 hours, with a therapeutic goal of 53 to 73 seconds. Once the activated partial thromboplastin time was therapeutic, CRRT was then restarted with the standard protocol of RCA with an acid citrate dextrose solution A rate of 300 mL/min. No more ECC clotting was observed, and after 12 hours of CRRT, the patient’s blood pressure improved, and he was weaned off intravenous norepinephrine. The patient then remained on CRRT for the life of the filter. After 70 hours, the filter was changed and CRRT restarted. However, at that time, the patient was found to have a deep venous thrombosis. Partly driven by financial considerations, the argatroban infusion was discontinued, and the patient was placed on a heparin infusion to treat his deep venous thrombosis. CRRT was then maintained for the next 42 hours until the patient was transitioned to intermittent HD. Despite recovery of renal function and resolution of septic shock, the patient continued to have persistent fungal mediastinitis and was eventually discharged to home hospice.
DISCUSSION Argatroban directly inhibits free and clot-bound thrombin and does not react with heparin antibodies.7 It is cleared by the liver, and its half-life is about 35 minutes in chronic dialysis.8 Its use as an anticoagulant in CRRT so far has been mainly limited to patients with heparin-induced thrombocytopenia or AT3 deficiency.3–6 In humans, prediction algorithms have been developed to estimate argatroban dose requirements during CRRT.4 Although this patient had been on subcutaneous hep-
arin before initiation of CRRT, his platelet counts remained relatively stable throughout his hospital course suggesting that heparin-induced thrombocytopenia was unlikely. Inherited AT3 deficiency is a rare autosomal dominant disorder with an estimated prevalence in healthy subjects of 0.03% to 0.05%.9 It is initially treated with intravenous unfractionated heparin. However, some patients with AT3 deficiency are refractory to heparin, and thus in Japan, argatroban is approved for the prevention of ECC clotting in HD patients who have inherited AT3 deficiency.10 Given that this patient had no history or family history of recurrent veno-occlusive disease and AT3’s low prevalence, we believe that inherited AT3 deficiency in this case is not likely. AT3 plays a critical role in the regulation of coagulation, being the major inhibitor of thrombin (factor IIa), and is estimated to provide 80% of the natural anticoagulant effect against thrombin.11 The extrinsic coagulation pathway is activated in patients with severe sepsis and septic shock, and in these patients, significantly low AT3 concentrations have been observed.12 Thus, it is likely that an acquired AT3 deficiency can occur in patients with severe sepsis11 putting them at a higher risk for ECC clotting during renal replacement therapy. Because heparins exert their anticoagulant effects by increasing the efficacy of AT3, a patient with severe sepsis could be resistant to the anticoagulant effects of heparins. In this situation, argatroban, being a direct thrombin inhibitor, would provide adequate anticoagulation independent of AT3 levels. We believe that this could provide an explanation as to why this patient’s ECC clotting was resistant to RCA and heparin and subsequently resolved with the use of argatroban. Currently, our institutional protocol does not allow for CRRT in the absence of RCA. This is the main reason we chose to continue RCA in addition to argatroban. RCA interferes with multiple steps of the coagulation cascade by lowering ionized calcium, an important cofactor in many of the steps of the cascade. Literature suggests that ionized calcium can also potentiate the inhibition of thrombin by AT3.13 Thus, it is possible that if the AT3 levels were already low, that further lowering the ionized calcium concentrations with RCA could lead to a further decrease in AT3 activity resulting in a hypercoagulable state that was resistant to RCA. However, argatroban is an effective thrombin inhibitor regardless of the ionized calcium level.14 Argatroban may also have an independent value in patients with septic disseminated intravascular coagulation.15 In animal studies, argatroban has been demonstrated to improve in vivo microcirculation and reduce leukocyte adherence in septic rats.16 Our patient had quick resolution of his septic shock after being placed on an argatroban infusion and
TABLE 1. Summary of ECC clotting times with the various anticoagulation methods CRRT restart RCA (mL/hr) Prefilter heparin (U/hr) Argatroban (mg/kg/min) 0 1 2 3a 4 5
300 300 300 300 350 300
500 500 1000 0.5
ECC clot time (hr)
Platelets (K/mm3)
1.0 4.0 1.0 4.0 0.5 70.0b
269 336 336 260 260 260
a
Hemodialysis catheter was changed over guidewire before this restart. Seventy-two hours is the life of the filter. CRRT, continuous renal replacement therapy; ECC, extracorporeal circuit; RCA, regional citrate anticoagulation.
b
Ó 2012 Lippincott Williams & Wilkins
257
Ferguson et al
undergoing adequate CRRT, suggesting that argatroban could play an additional role in patients with septic shock. An important limitation of our study is that we did not measure levels of AT3, protein S, protein C, homocysteine and anticardiolipin antibodies. Nor did we check for a factor V Leiden mutation in our patient. However, the interpretation of these test results during critical illness would have been challenging and may not have changed the immediate management of his renal replacement therapy. In summary, we describe a case of recurrent ECC clotting in a patient with septic shock due to S prolificans infection, despite escalating RCA and adding heparin. ECC clotting only resolved after the use of argatroban as the primary anticoagulant. We believe that argatroban should be further investigated as an anticoagulant for CRRT in patients with septic shock. REFERENCES 1. Brophy PD, Somers MJG, Baum MA, et al. Multi-centre evaluation of anticoagulation in patients receiving continuous renal replacement therapy (CRRT). Nephrol Dial Transplant 2005;20:1416–21. 2. Tolwani AJ, Willie KM. Anticoagulation for continuous renal replacement therapy. Semin Dial 2009;22:141–5. 3. Ota K, Akizawa T, Hirasawa Y, et al. Effects of argatroban as an anticoagulant for haemodialysis in patients with antithrombin III deficiency. Nephrol Dial Transplant 2003;18:1623–30. 4. Link A, Girndt M, Selejan S, et al. Argatroban for anticoagulation in continuous renal replacement therapy. Crit Care Med 2009;37:105–10. 5. Tang IY, Cox DS, Patel K, et al. Argatroban and renal replacement therapy in patients with heparin-induced thrombocytopenia. Ann Pharmacother 2005;39:231–6.
258
6. Reddy BV, Grossman EJ, Trevino SA, et al. Argatroban anticoagulation in patients with heparin-induced thrombocytopenia requiring renal replacement therapy. Ann Pharmacother 2005;39:1601–5. 7. Okamoto S, Hijikata-Okunomiya A. Synthetic selective inhibitors of thrombin. Methods Enzymol 1993;222:328–40. 8. Murray PT, Reddy BV, Grossman EJ, et al. A prospective comparison of three argatroban treatment regimens during hemodialysis in endstage renal disease. Kidney Int 2004;66:2446–53. 9. Tait RC, Walker ID, Perry DJ, et al. Prevalence of antithrombin deficiency in the healthy population. Br J Haemtol 1994;87:106–12. 10. Hara T, Naito K. Inherited antithrombin deficiency and end stage renal disease. Med Sci Monit 2005;11:346–54. 11. Maclean PS, Tait RC. Hereditary and acquired antithrombin deficiency. Drugs 2007;67:1429–40. 12. Gando S, Nanzaki S, Sasaki S, et al. Activation of the extrinsic coagulation pathway in patients with severe sepsis and septic shock. Crit Care Med 1998;26:2005–9. 13. Long WF, Williamson FB. Potentiation by calcium ions of the antithrombin III inhibition of thrombin. Biochem Biophys Res Commun 1982;104:363–8. 14. Bush LR. Argatroban, a selective, potent thrombin inhibitor. Cardiovasc Drug Rev 1991;9:247–63. 15. Beyer J, Halbritter K, Weise M, et al. Influence of antithrombin and argatroban on disseminated intravascular coagulation parameters in a patient with septic shock. Thromb Res 2009;124:383–6. 16. Fuchs C, Ladwig E, Zhou J, et al. Argatroban administration reduces leukocyte adhesion and improves capillary perfusion within the intestinal microcirculation in experimental sepsis. Thromb Haemost 2010; 104:1022–8.
Volume 345, Number 3, March 2013
Recurring Extracorporeal Circuit Clotting During Continuous Renal Replacement Therapy in Fungal Sepsis: Successful Treatment With Argatroban Lee M. Ferguson, MD, Albert W. Dreisbach, MD, Éva Csongrádi, MD, Luis A. Juncos, MD and Tibor Fulop, MD
Abstract: The relative effectiveness of anticoagulation strategies during continuous renal replacement therapy (CRRT) may vary according to the clinical circumstances. In this study, the case of a 46-year-old man who developed fungal mediastinitis with the pathogen Scedosporium prolificans after coronary bypass surgery is reported. Numerous debridements and multiple antifungal agents were not effective in this patient. Miltefosine, a non-Food and Drug Administration-approved agent, was started after institutional review board request and approval. CRRT was initiated with regional citrate anticoagulation (RCA) for clinical sepsis with acute kidney injury. Subsequently, crescendo clotting of the extracorporeal circuit (ECC) occurred. Multiple interventions, including escalating RCA, adding increasing heparin to RCA and exchanging the dialysis catheter, were not effective. Argatroban anticoagulation was started without further ECC clotting, and the patient recovered from both acute kidney injury and septic shock, despite continued miltefosine administration. Sepsis may contribute to recurrent ECC clotting. Argatroban, a direct thrombin inhibitor, had a disproportionate effectiveness to maintain ECC patency in this patient. Key Indexing Terms: Argatroban; Continuous renal replacement therapy; Miltefosine; Regional citrate anticoagulation; Sepsis. [Am J Med Sci 2013;345(3):256–258.]
I
INTRODUCTION
n critically-ill patients, extracorporeal circuit (ECC) clotting is a frequent complication of continuous renal replacement therapy (CRRT). Traditionally, this is prevented by using regional citrate anticoagulation (RCA) or prefilter unfractionated heparin.1,2 Argatroban is a direct thrombin inhibitor whose role in CRRT anticoagulation has been limited to patients with either heparin-induced thrombocytopenia or antithrombin III (AT3) deficiency.3–6 Here, we report a case of recurrent ECC clotting in a patient with fungal sepsis, despite RCA and the addition of unfractionated heparin. After switching to argatroban as the primary anticoagulant, the patient was able to undergo CRRT without further ECC clotting.
CASE REPORT Our patient is a 46-year-old Caucasian man with type II diabetes mellitus, hypertension and coronary artery disease who underwent a 4-vessel coronary artery bypass graft after having a myocardial infarction. Postoperatively, the patient developed From the Department of Medicine (LMF) and Division of Nephrology, Department of Medicine (AWD, LAJ, TF), University of Mississippi Medical Center, Jackson, Mississippi; First Department of Medicine (ÉC), Medical and Health Science Center, University of Debrecen, Debrecen, Hungary; and Department of Physiology and Biophysics (LAJ), University of Mississippi Medical Center, Jackson, Mississippi. Submitted June 14, 2012; accepted in revised form August 27, 2012. The authors have no financial or other conflicts of interest to disclose. Correspondence: Lee M. Ferguson, MD, Department of Medicine, University of Mississippi Medical Center, 2500 North State Street, L 504, Jackson, MS 39216-4505 (E-mail: [email protected]).
256
sternal osteomyelitis, with sternal cultures growing an unusual fungal pathogen, Scedosporium prolificans. Over the next year, the patient continued to have recurring fungal sternal osteomyelitis, despite multiple courses of various intravenous antifungal medications and numerous sternal debridements (48 in all). He was subsequently left with a complete sternotomy with definitive closure by omental flap and skin grafting. Despite all these measures, he experienced yet another recurrence of fungal osteomyelitis and mediastinitis. Therefore, after obtaining institutional review board approval, the patient was placed on the experimental antifungal drug, miltefosine (MILT). Two days after initiation of MILT, he developed hypotension, clinical sepsis and acute kidney injury (AKI), necessitating temporary hemodialysis (HD). MILT was thought to have contributed to his AKI and was discontinued during the admission. He was subsequently diagnosed with gangrenous cholecystitis, necessitating emergency surgical cholecystectomy; he then gradually recovered. Two months later, the patient was readmitted with purulent drainage from his sternal graft site. The initial physical examination revealed a temperature of 98.6⁰F, a heart rate of 72 beats per minute, a blood pressure of 96/54 mm of Hg, a respiratory rate of 18 breaths per minute and an O2 saturation of 97%. The sternal graft site had erythema around the edges and edema of the flap. Bilateral lower extremities had moderate pitting edema. The initial laboratory values were as follows: white blood cell count, 5,900 per cubic millimeter; hematocrit, 22.9%; platelet count, 170,000 per cubic millimeter; glucose level, 59 mg/dL; sodium level, 135 mmol/L; potassium level, 4.4 mmol/L; chloride level, 99 mmol/L; bicarbonate level, 29 mmol/L; blood urea nitrogen level, 46 mg/dL; and creatinine level, 2.24 mg/dL. The patient was placed on voriconazole and caspofungin for presumed recurrent fungal mediastinitis and daptomycin for possible methicillin-resistant Staphylococcus aureus infection. Additionally, oral MILT was reintroduced to his regimen. Over the next 2 days, the patient became oliguric and developed volume overload with worsening pulmonary edema. Given that the patient had a rising serum creatinine, he was again diagnosed with AKI complicating his chronic kidney disease. A bilateral renal ultrasound revealed increased cortical echogenicity of both kidneys and no evidence of hydronephrosis. Despite aggressive diuretic therapy, the patient continued to have worsening volume overload and was transferred to the medical intensive care unit for the initiation of CRRT. At the time of transfer, serum creatinine level measured 4.4 mg/dL; the blood urea nitrogen level was 63 mg/dL and platelets were 269,000 per cubic millimeter. A temporary right internal jugular HD catheter was placed, and CRRT was started with RCA. CRRT used an acid citrate dextrose solution A at a prefilter flow rate of 300 mL/ hr, a blood flow rate of 200 mL/min, a hemofiltration rate of 2500 mL/hr with premixed calcium-free replacement fluid
The American Journal of the Medical Sciences
Volume 345, Number 3, March 2013
Argatroban During CRRT
(split 50%–50% between prefilter and postfilter dilution), a systemic calcium chloride infusion, and an effective filtration fraction of 11% during the procedure. Initiation of CRRT was complicated by hypotension, and the patient was subsequently given a 700-mL bolus of intravenous normal saline and started on an intravenous norepinephrine infusion for septic shock. CRRT was started and lasted for approximately 1 hour before ECC clotting occurred. Blood clots were found in CRRT tubing and HD catheter ports. Despite escalating RCA and adding prefilter heparin, the patient continued to have recurrent ECC clotting. Table 1 shows the ECC clotting times with the various anticoagulation methods. Fibrinogen levels were only mildly elevated at 401 and 485 mg/dL (reference range, 180–400 mg/dL). Serum procalcitonin level was markedly elevated at 15.81 ng/mL (reference range, 0.00–0.05 ng/mL). After multiple failed restarts of CRRT secondary to ECC clotting, the patient was placed on a critical illnessadjusted dose of argatroban of 0.5 mg/kg/min. Activated partial thromboplastin times were checked, initially every 2 hours, with a therapeutic goal of 53 to 73 seconds. Once the activated partial thromboplastin time was therapeutic, CRRT was then restarted with the standard protocol of RCA with an acid citrate dextrose solution A rate of 300 mL/min. No more ECC clotting was observed, and after 12 hours of CRRT, the patient’s blood pressure improved, and he was weaned off intravenous norepinephrine. The patient then remained on CRRT for the life of the filter. After 70 hours, the filter was changed and CRRT restarted. However, at that time, the patient was found to have a deep venous thrombosis. Partly driven by financial considerations, the argatroban infusion was discontinued, and the patient was placed on a heparin infusion to treat his deep venous thrombosis. CRRT was then maintained for the next 42 hours until the patient was transitioned to intermittent HD. Despite recovery of renal function and resolution of septic shock, the patient continued to have persistent fungal mediastinitis and was eventually discharged to home hospice.
DISCUSSION Argatroban directly inhibits free and clot-bound thrombin and does not react with heparin antibodies.7 It is cleared by the liver, and its half-life is about 35 minutes in chronic dialysis.8 Its use as an anticoagulant in CRRT so far has been mainly limited to patients with heparin-induced thrombocytopenia or AT3 deficiency.3–6 In humans, prediction algorithms have been developed to estimate argatroban dose requirements during CRRT.4 Although this patient had been on subcutaneous hep-
arin before initiation of CRRT, his platelet counts remained relatively stable throughout his hospital course suggesting that heparin-induced thrombocytopenia was unlikely. Inherited AT3 deficiency is a rare autosomal dominant disorder with an estimated prevalence in healthy subjects of 0.03% to 0.05%.9 It is initially treated with intravenous unfractionated heparin. However, some patients with AT3 deficiency are refractory to heparin, and thus in Japan, argatroban is approved for the prevention of ECC clotting in HD patients who have inherited AT3 deficiency.10 Given that this patient had no history or family history of recurrent veno-occlusive disease and AT3’s low prevalence, we believe that inherited AT3 deficiency in this case is not likely. AT3 plays a critical role in the regulation of coagulation, being the major inhibitor of thrombin (factor IIa), and is estimated to provide 80% of the natural anticoagulant effect against thrombin.11 The extrinsic coagulation pathway is activated in patients with severe sepsis and septic shock, and in these patients, significantly low AT3 concentrations have been observed.12 Thus, it is likely that an acquired AT3 deficiency can occur in patients with severe sepsis11 putting them at a higher risk for ECC clotting during renal replacement therapy. Because heparins exert their anticoagulant effects by increasing the efficacy of AT3, a patient with severe sepsis could be resistant to the anticoagulant effects of heparins. In this situation, argatroban, being a direct thrombin inhibitor, would provide adequate anticoagulation independent of AT3 levels. We believe that this could provide an explanation as to why this patient’s ECC clotting was resistant to RCA and heparin and subsequently resolved with the use of argatroban. Currently, our institutional protocol does not allow for CRRT in the absence of RCA. This is the main reason we chose to continue RCA in addition to argatroban. RCA interferes with multiple steps of the coagulation cascade by lowering ionized calcium, an important cofactor in many of the steps of the cascade. Literature suggests that ionized calcium can also potentiate the inhibition of thrombin by AT3.13 Thus, it is possible that if the AT3 levels were already low, that further lowering the ionized calcium concentrations with RCA could lead to a further decrease in AT3 activity resulting in a hypercoagulable state that was resistant to RCA. However, argatroban is an effective thrombin inhibitor regardless of the ionized calcium level.14 Argatroban may also have an independent value in patients with septic disseminated intravascular coagulation.15 In animal studies, argatroban has been demonstrated to improve in vivo microcirculation and reduce leukocyte adherence in septic rats.16 Our patient had quick resolution of his septic shock after being placed on an argatroban infusion and
TABLE 1. Summary of ECC clotting times with the various anticoagulation methods CRRT restart RCA (mL/hr) Prefilter heparin (U/hr) Argatroban (mg/kg/min) 0 1 2 3a 4 5
300 300 300 300 350 300
500 500 1000 0.5
ECC clot time (hr)
Platelets (K/mm3)
1.0 4.0 1.0 4.0 0.5 70.0b
269 336 336 260 260 260
a
Hemodialysis catheter was changed over guidewire before this restart. Seventy-two hours is the life of the filter. CRRT, continuous renal replacement therapy; ECC, extracorporeal circuit; RCA, regional citrate anticoagulation.
b
Ó 2012 Lippincott Williams & Wilkins
257
Ferguson et al
undergoing adequate CRRT, suggesting that argatroban could play an additional role in patients with septic shock. An important limitation of our study is that we did not measure levels of AT3, protein S, protein C, homocysteine and anticardiolipin antibodies. Nor did we check for a factor V Leiden mutation in our patient. However, the interpretation of these test results during critical illness would have been challenging and may not have changed the immediate management of his renal replacement therapy. In summary, we describe a case of recurrent ECC clotting in a patient with septic shock due to S prolificans infection, despite escalating RCA and adding heparin. ECC clotting only resolved after the use of argatroban as the primary anticoagulant. We believe that argatroban should be further investigated as an anticoagulant for CRRT in patients with septic shock. REFERENCES 1. Brophy PD, Somers MJG, Baum MA, et al. Multi-centre evaluation of anticoagulation in patients receiving continuous renal replacement therapy (CRRT). Nephrol Dial Transplant 2005;20:1416–21. 2. Tolwani AJ, Willie KM. Anticoagulation for continuous renal replacement therapy. Semin Dial 2009;22:141–5. 3. Ota K, Akizawa T, Hirasawa Y, et al. Effects of argatroban as an anticoagulant for haemodialysis in patients with antithrombin III deficiency. Nephrol Dial Transplant 2003;18:1623–30. 4. Link A, Girndt M, Selejan S, et al. Argatroban for anticoagulation in continuous renal replacement therapy. Crit Care Med 2009;37:105–10. 5. Tang IY, Cox DS, Patel K, et al. Argatroban and renal replacement therapy in patients with heparin-induced thrombocytopenia. Ann Pharmacother 2005;39:231–6.
258
6. Reddy BV, Grossman EJ, Trevino SA, et al. Argatroban anticoagulation in patients with heparin-induced thrombocytopenia requiring renal replacement therapy. Ann Pharmacother 2005;39:1601–5. 7. Okamoto S, Hijikata-Okunomiya A. Synthetic selective inhibitors of thrombin. Methods Enzymol 1993;222:328–40. 8. Murray PT, Reddy BV, Grossman EJ, et al. A prospective comparison of three argatroban treatment regimens during hemodialysis in endstage renal disease. Kidney Int 2004;66:2446–53. 9. Tait RC, Walker ID, Perry DJ, et al. Prevalence of antithrombin deficiency in the healthy population. Br J Haemtol 1994;87:106–12. 10. Hara T, Naito K. Inherited antithrombin deficiency and end stage renal disease. Med Sci Monit 2005;11:346–54. 11. Maclean PS, Tait RC. Hereditary and acquired antithrombin deficiency. Drugs 2007;67:1429–40. 12. Gando S, Nanzaki S, Sasaki S, et al. Activation of the extrinsic coagulation pathway in patients with severe sepsis and septic shock. Crit Care Med 1998;26:2005–9. 13. Long WF, Williamson FB. Potentiation by calcium ions of the antithrombin III inhibition of thrombin. Biochem Biophys Res Commun 1982;104:363–8. 14. Bush LR. Argatroban, a selective, potent thrombin inhibitor. Cardiovasc Drug Rev 1991;9:247–63. 15. Beyer J, Halbritter K, Weise M, et al. Influence of antithrombin and argatroban on disseminated intravascular coagulation parameters in a patient with septic shock. Thromb Res 2009;124:383–6. 16. Fuchs C, Ladwig E, Zhou J, et al. Argatroban administration reduces leukocyte adhesion and improves capillary perfusion within the intestinal microcirculation in experimental sepsis. Thromb Haemost 2010; 104:1022–8.
Volume 345, Number 3, March 2013