Recombinant Factor Seven Therapy for Postoperative Bleeding in Neonatal and Pediatric Cardiac Surgery

Transcription

1 Recombinant Factor Seven Therapy for Postoperative Bleeding in Neonatal and Pediatric Cardiac Surgery Hemant S. Agarwal, MBBS, Jo E. Bennett, BS, Kevin B. Churchwell, FAAP, MD, Karla G. Christian, MD, Davis C. Drinkwater Jr, MD, Yi He, MS, and Mary B. Taylor, MD Department of Pediatrics, Division of Pediatric Critical Care, Department of Pediatric Cardiothoracic Surgery, and Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee Background. Severe bleeding is a major complication in the postoperative pediatric cardiac surgery patients. We evaluated the efficacy and safety of recombinant factor seven (rfviia) therapy in this patient population. Methods. A retrospective unmatched case-control study for the previous five years in a single institution was undertaken. Patients with severe bleeding treated with rfviia therapy (study group) were compared with patients treated with blood products only (control group) using analysis of variance. Mediastinal bleeding, blood products transfusion, and coagulation studies before and six hours after the first dose of rfviia therapy were analyzed using the Student paired t test. The dose, frequency, and side-effects of rfviia therapy were studied. Results. Forty-six patients with severe bleeding were studied. Twenty-three of 24 patients in the study group, including 12 patients placed on extracorporeal membrane oxygenation (ECMO), responded to rfviia therapy (mean dose g/kg/dose). There was significant reduction in chest tube drainage (from ml/kg/hour to ml/kg/hour, p ) along with significant reduction of blood products transfusion (p < 0.001) in the study group patients as compared with control group patients. One patient who failed to respond had surgical bleeding. Two patients developed major thrombotic complications that included clots in the ECMO circuit and thrombosis at bleeding arterial line site resulting in limb ischemia. Four additional patients in the study group developed mediastinal clots. Overall, 25% of patients developed thrombosis after rfviia therapy. Conclusions. The rfviia therapy seems to be an effective treatment for severe bleeding in postoperative pediatric cardiac surgery patients in the absence of surgical bleeding. It must be judiciously used in patients bleeding from multiple sites or having preexistent clots in the ECMO circuit to prevent major thrombotic complications. (Ann Thorac Surg 2007;84:161 9) 2007 by The Society of Thoracic Surgeons Postoperative bleeding is a major complication in neonates and infants undergoing cardiac surgery. Mediastinal blood loss as high as 15 to 110 ml/kg has been reported in some reviews [1]. Limited therapeutic options are available to treat severe bleeding. Blood products, including fresh frozen plasma, cryoprecipitate, and pooled platelets, are administered to correct the coagulopathy and packed red blood cells are concurrently given to correct the anemia. Administration of large amounts of blood product transfusion has the risk of transfusion-related reactions and disease transmission. Alternative therapies like antifibrinolytic drugs, including tranexamic acid, aprotinin, and epsilonaminocaproic acid, are more effective prophylactically than for treatment of postoperative bleeding [2]. Reexploration of the chest in patients with severe bleeding reveals a surgically manageable source of bleeding in less Accepted for publication Feb 20, Address correspondence to Dr Taylor, Vanderbilt Children s Hospital, 2200 Children s Way, 5121 B Doctors Office Tower, Nashville, TN ; mary.b.taylor@vanderbilt.edu. than 50% of cases [3]. Thus, an effective and safe medical therapy needs to be studied for these patients. Recombinant factor VIIa (rfviia, Novoseven; Novo Nordisk, Princeton, NJ) was introduced in the 1980s to treat bleeding in hemophilia patients [4]. It is currently being used as an off-label drug for treatment of severe bleeding in clinical conditions such as trauma, adult cardiac surgery, and liver transplantation [5 7]. Case reports and limited case series favor the role of rfviia therapy in postoperative pediatric cardiac surgery patients [8 12] although a large study is still lacking. We reviewed our five years experience of rfviia therapy for severe postoperative bleeding, including patients placed on extracorporeal membrane oxygenation (ECMO) to study the efficacy and safety of rfviia. Material and Methods This retrospective study was conducted in the pediatric cardiac intensive care unit (PCICU) of the Monroe Carell Jr. Children s Hospital at Vanderbilt. An approval from 2007 by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 162 AGARWAL ET AL Ann Thorac Surg RECOMBINANT FACTOR SEVEN THERAPY 2007;84:161 9 Table 1. Preoperative and Immediate Postoperative Details of Patients Characteristics Study Group (n 24) Control Group (n 22) Demographic details: Age (days) 9.5 a (4 3,285) 7 a (2 240) Weight (kgs) 3.5 a (2.4 51) 3.2 a (2 6.7) Cardiac defect: HLHS 12 8 TGA 6 6 AS 2 2 BDG 1 2 OTHERS 3 5 ECMO a Median; range in parentheses. AS aortic stenosis; BDG bidirectional Glenn shunt; ECMO extracorporeal membrane oxygenation; HLHS hypoplastic left heart syndrome; TGA transposition of great arteries. the Institutional Review Board was obtained including waiver of consent authorization. Medical charts of all patients who had severe bleeding in the immediate postoperative period after cardiac surgery from January 2000 to December 2004 were reviewed. Children less than 19 years of age who had no known bleeding disorder were included in the study. Patients who received rfviia in the immediate postoperative period for severe bleeding were categorized to a study group and patients who were managed with blood products only to control their bleeding were categorized to the control group. Demographic characteristics such as age, weight, sex, and preoperative cardiac defect were noted. Operative records were reviewed for cardiopulmonary bypass (CPB) time, aortic cross-clamp time, and the surgical procedures performed. Patients placed on ECMO postoperatively for cardiopulmonary support were also included in the study. Any patient in the control group who responded to surgical exploration for severe bleeding was excluded from the control group. For this retrospective review, selection of patients with severe bleeding was defined in each patient based on mediastinal chest tube blood loss. At the completion of surgery after being weaned off the CPB, all patients were monitored on an hourly basis for chest tube bleeding. Mediastinal bleeding, more than 10 ml/kg per hour in neonates and infants or more than 100 ml/hour in adolescents for at least one hour time period, was used to identify patients for inclusion in the study group. Hematology laboratory tests including prothrombin time (PT), activated partial thromboplastin time (PTT), international normalized ratio (INR), platelet count, fibrinogen levels, and hematocrit were evaluated. Bleeding and anemia were empirically controlled by the attending physician by administration of blood products that generally included the following: platelets, 1 unit/10 kg; fresh frozen plasma, 10 to 15 ml/kg; cryoprecipitate, 1 unit/5 kg; and packed red cells, 10 to 15 ml/kg. Further blood products administration was based on laboratory results of INR, PT, PTT, fibrinogen levels, platelet count, and hematocrit values. Surgical reexploration was constantly assessed by the attending physician in consultation with the surgeon. Administration of rfviia at this time period was based on the discretion of the attending physician. Because rfviia was employed as rescue therapy for critically ill patients and there were no experimental interventions, informed consent was deemed unnecessary. Once decision was made to give rfviia, a bolus injection was administered and the dose given was at the discretion of the attending physician. Mediastinal chest tube drainage and the requirement of blood products administration was reassessed after rfviia therapy. Patients did not receive any further rfviia if there was significant reduction in the chest tube drainage that was sustained over the next two to four hours. However, if the chest tube drainage did not show a sustained downward trend over the next few hours, coagulation studies and hematocrit values were repeated and blood products were administered. Further administrations of rfviia doses and (or) mediastinal reexploration were at the discretion of the attending physician and the surgeon. The safety of the drug was monitored in all the patients who received rfviia therapy. Complications known after rfviia therapy, especially thrombosis of blood vessels in the patient or thrombosis in the ECMO circuit, were specifically assessed. Thrombosis of the ECMO circuit was assessed by visual inspection and changes in the pressure gradients in the ECMO circuit tubing. Thrombosis of blood vessels in the patients were assessed by clinical examination of the peripheral pulses, bedside portable Doppler testing of the pulses, and distal tissue perfusion. Mediastinal clots were assessed by lack of chest tube drainage associated with distention of the mediastinum in the delayed sternal closure patients. Mediastinal chest tube bleeding and volume of blood products administered, including fresh frozen plasma, cryoprecipitate, platelets, and packed red cells given to the patient, and laboratory hematology results were analyzed in all patients using the statistical software SAS (SAS Institute Inc, Cary, NC). All continuous results are expressed as mean standard deviation. Analysis of variance (ANOVA) was used to compare data in the study group patients before, and for six hours after, Table 2. Recombinant Factor Seven (rfviia) Therapy Administered to the Study Group Patients Number of rfviia Doses Amount of rfviia Dose ( g/kg) Dose Interval (hours) Response to rfviia 1(n 15) OR to 1 st dose: Yes (n 7) st to 2 nd dose: Yes (n 1) nd to 3 rd dose: 2 Yes 4(n 1) rd to 4 th dose: 2 No n number of patients; OR operating room.

3 Ann Thorac Surg AGARWAL ET AL 2007;84:161 9 RECOMBINANT FACTOR SEVEN THERAPY Table 3. Comparison of Chest Tube Bleeding, Blood Products Administration, and Laboratory Hematology Values in Study Group (Before and After Recombinant Factor Seven [rfviia] Therapy) and Control Group Patients (Without rfviia Therapy) in a Similar Postoperative Time Period Study Group (n 23) Control Group (n 22) 163 Before rfviia After rfviia 0 4 Hours 5 10 Hours Mediastinal/chest tube bleeding (ml/kg/hr) b Blood product administration: Packed red cell transfusion (ml/kg) b Fresh frozen plasma (ml/kg) b Platelets (ml/kg) b Cryoprecipitate (ml/kg) b Hematology values: Packed cell volume (%) c Platelet count (thousands/ L) c INR c d d PT (seconds) c d d PTT (seconds) c d Fibrinogen levels (mg/dl) c d a Mean standard deviation; b p 0.001; c n 19 patients; d p 0.05 (p value based on paired t test comparing before and after rfviia therapy in study group as well as comparing 0 to 4 hours after CPB and 5 to 10 hours after cardiopulmonary bypass in control group). INR international normalized ratio; n number of patients; PT prothrombin time; PTT activated partial thromboplastin time. the first dose of rfviia versus data in the control group in the initial period (0 to 4 hours after surgery) and later period (5 to 10 hours after surgery). We selected this time period in the control group as the mean time of administration of the first dose of rfviia in the study group patients was 3.9 hours after surgery. In the study group patients, data were also analyzed before and after the first dose of rfviia using the Student paired t test. A 2 or Fisher exact test was used for categoric comparisons. Results Forty-six children with severe bleeding in the immediate postoperative period of cardiac surgery were identified over a period of five years. Twenty-four patients in the study group received rfviia therapy and 22 patients in the control group received only blood products. Demographic details, including the age, weight, and the cardiac defect of patients in the study group and the control group are shown in Table 1. Twelve patients in the study group and 15 patients in the control group were placed on ECMO for low cardiac output. The ECMO flow was maintained in all patients at approximately 100 to 150 ml/kg per minute. A continuous infusion of heparin was used on all ECMO patients to maintain an activated clotting time (ACT) of 180 to 200 seconds. No antifibrinolytic drugs were used in any patient. The rfviia therapy was administered in the study group patients after a mean time of 3.9 hours after CPB. The dose, frequency, and time interval between rfviia doses are shown in Table 2. Fifteen of the 24 patients (62%) received one dose of rfviia, seven patients (29%) received two doses of rfviia, and one patient each received three and four doses of rfviia, respectively. Collectively, the mean dose of rfviia administered to all patients was g/kg. Patients requiring ECMO support received smaller ( g/kg vs g/kg) and frequent ( vs ) doses of rfviia as compared with patients not placed on ECMO, although not statistically significant. Fig 1. Comparison of chest tube bleeding and blood products administration in study group and control group patients in a similar postoperative time period. Study group: A, before rfviia therapy; B, after rfviia therapy; control group: C, 0 to 4 hours after surgery; D, 5 to 10 hours after surgery; (*p 0.05 bars displayed as mean SD; Cryo cryoprecipitate; CT bleed chest bleeding; FFP fresh frozen plasma; PRBCs packed red blood cells transfusion.)

4 164 AGARWAL ET AL Ann Thorac Surg RECOMBINANT FACTOR SEVEN THERAPY 2007;84:161 9 Table 4. Chest Tube Bleeding, Blood Products Administration, and Laboratory Hematology Values in Responsive Study Group (Before and After Recombinant Factor Seven [rfviia] Therapy) and Control Group Patients (Without rfviia Therapy) on Extracorporeal Membrane Oxygenation In a Similar Postoperative Time Period Study Group (n 11) Control Group (n 15) Before rfviia After rfviia 0 4 Hours 5 10 Hours Mediastinal/chest tube bleeding (ml/kg/hour) b Blood product administration: Packed red cell transfusion (ml/kg) b b Fresh frozen plasma (ml/kg) b b Platelets (ml/kg) Cryoprecipitate (ml/kg) b Hematology values: Packed cell volume (%) Platelet count (thousands/ L) INR c b PT (seconds) b c PTT (seconds) b Fibrinogen levels (mg/dl) b a Mean standard deviation; b p 0.05; c p (p value based on paired t test comparing before and after rfviia therapy in study group as well as comparing 0 to 4 hours after CPB and 5 to 10 hours after CPB in control group). CPB cardiopulmonary bypass; INR international normalized ratio; n number of patients; PT prothrombin time; PTT activated partial thromboplastin time. Twenty-three of 24 patients in the study group responded to rfviia therapy (96%) as evidenced by a significant reduction in bleeding from chest tubes and requirement of blood products. One patient who did not respond to four doses of rfviia had a surgical cause of bleeding on mediastinal reexploration by the surgeon. In the 23 responsive patients there was a significant reduction in chest tube drainage from ml/kg per hour to ml/kg per hour before, and for six hours after, administration of the first dose of rfviia (p ) (Table 3). There was a simultaneous significant reduction in transfusion of blood products, including packed red blood cells (p 0.004), pooled platelets (p ), fresh frozen plasma (p ), and cryoprecipitate (p ) (Table 3). Analyzing chest tube bleeding and blood products administration in the study group and control group patients with ANOVA, and adjusting for age, weight, and ECMO use, there was a significant reduction in chest tube bleeding (p ) in association with a reduction in amount of packed red cells transfusion (p 0.001), platelets transfusion (p ), fresh frozen plasma transfusion (p ), and cryoprecipitate transfusion (p 0.008) (Fig 1). The subjects were also analyzed based on requirement of ECMO for low cardiac output. Twelve patients in the study group and 15 patients in the control group were placed on ECMO. Eleven of the 12 patients (92%) in the study group on ECMO responded to rfviia therapy. There was a significant reduction in chest tube bleeding (p ) along with a significant reduction in administration of blood products in the 11 responsive study group patients (Table 4, Fig 2). The control group patients on ECMO had no reduction in chest tube bleeding over time and required significantly more packed red cells transfusion (p 0.035) and fresh frozen plasma (p 0.016) over time for their bleeding (Table 4). Twelve patients in the study group and seven patients in the control group did not require ECMO support. All 12 Fig 2. Comparison of chest tube bleeding and blood products administration in study group and control group patients on extracorporeal membrane oxygenation in a similar postoperative time period. Study group: A, before rfviia therapy; B, after rfviia therapy; control group: C, 0 to 4 hours after surgery; D, 5 to 10 hours after surgery (*p 0.05; bars displayed as mean SD; Cryo cryoprecipitate; CT bleed chest bleeding; FFP fresh frozen plasma; PRBCs packed red blood cells transfusion.)

5 Ann Thorac Surg AGARWAL ET AL 2007;84:161 9 RECOMBINANT FACTOR SEVEN THERAPY Table 5. Chest Tube Bleeding, Blood Products Administration and Laboratory Hematology Values in Study Group (Before and After Recombinant Factor Seven [rfviia] Therapy) and Control Group Patients (Without rfviia Therapy) Not Requiring Extracorporeal Membrane Oxygenation Support in a Similar Postoperative Time Period Study Group (n 12) Control Group (n 7) 165 Before rfviia After rfviia 0 4 Hours 5 10 Hours Mediastinal/chest tube bleeding (ml/kg/hour): b Blood product administration: Packed red cell transfusion (ml/kg) b Fresh frozen plasma (ml/kg) b Platelets (ml/kg) Cryoprecipitate (ml/kg) b Hematology values: Packed cell volume (%) c Platelet count (thousands/ L) c INR c d b PT (seconds) c d d PTT (seconds) c b Fibrinogen levels (mg/dl) c b a Mean standard deviation; b p 0.05; c n 7 patients; d p (p value based on paired t test comparing before and after rfviia therapy in study group as well as comparing 0 to 4 hours after cardiopulmonary bypass and 5 to 10 hours after CPB in control group). CPB cardiopulmonary bypass; INR international normalized ratio; n number of patients; PT prothrombin time; PTT activated partial thromboplastin time. patients in the study group responded to rfviia therapy. There was a statistically significant reduction of chest tube bleeding (p ) and a significant reduction in amount of blood products administration after rfviia therapy in the study group as compared with control group patients (Table 5, Fig 3). As anticipated, patients on ECMO had more chest tube bleeding and required more blood products initially in both the study group and control group patients. Hematology laboratory studies in all the study group patients before and after the first doses of rfviia as well as in the control group patients are shown in Table 3. The coagulation studies could be evaluated in 19 study group patients as five patients received the first dose of rfviia in the operating room. There was a significant reduction in PT and INR after the first dose of rfviia in the study group patients (p 0.05). There was also a significant improvement in the PT, PTT, INR, and fibrinogen levels in the control group patients (p 0.05). A similar trend was noticed in both subsets of patients requiring ECMO or no ECMO support in the immediate postoperative period (Tables 4 and 5). Five patients in the study group underwent surgical reexploration for mediastinal chest tube bleeding (one patient identified to have a surgical cause of bleeding) as compared with eight patients in the control group needing surgical reexploration (p 0.15). Major thrombosis formation was noticed in two patients after administration of rfviia therapy. The first patient was a full-term neonate who underwent modified Norwood stage 1 repair with right ventricle to pulmonary artery conduit. He was placed on ECMO coming off CPB for low cardiac output. The ECMO circuit had normal pressures although microbubbles and a small clot were noticed in the top of the heater reservoir. He received one dose of 50 g/kg of rfviia four hours after the surgery. The bleeding significantly improved, although there was an elevation of preoxygenator and postoxygenator circuit Fig 3. Comparison of chest tube bleeding and blood products administration in study group and control group patients not placed on extracorporeal membrane oxygenation in a similar postoperative time period. Study group: A, before rfviia therapy; B, after rfviia therapy; control group: C, 0 to 4 hours after surgery; D, 5 to 10 hours after surgery. (*p 0.05; bars displayed as mean SD; CT bleed chest bleeding; PRBCs packed red blood cells transfusion; FFP fresh frozen plasma; Cryo cryoprecipitate.)

6 166 AGARWAL ET AL Ann Thorac Surg RECOMBINANT FACTOR SEVEN THERAPY 2007;84:161 9 pressure by 10 points without clinical consequences. Eight hours later there was marked increase in the preoxygenator and postoxygenator pressures secondary to presumed clots in the ECMO circuit. The circuit was emergently changed and the patient was resuscitated successfully. The patient was weaned off ECMO on day 3. The second patient was also a full-term neonate who underwent modified Norwood stage 1 repair with a right ventricle to pulmonary artery conduit. He was placed on ECMO coming off CPB for low cardiac output. He had significant bleeding from the mediastinal chest tubes as well as from the right subclavian central venous catheter insertion site. Multiple unsuccessful attempts were made to place a percutaneous femoral arterial line after the surgery. The patient started oozing from the attempted arterial site. He received two doses of rfviia of 40 g/kg per dose at two hour intervals. He developed poor perfusion of the lower limb, where an arterial line was attempted 12 hours after administration of the second dose of rfviia. An arterial duplex study revealed a dampened arterial flow pattern at femoral bifurcation with a loss of arterial flow at the proximal femoral artery consistent with arterial thrombosis. An echocardiograph on the same day revealed a 5 mm pericardial thrombus and a 3 mm atrial thrombus. The central venous catheter was replaced and the patient was weaned off ECMO the same day. He was placed on heparin infusion to prevent progression of the thrombosis. He subsequently underwent a left atrial thrombectomy, pericardial clot removal, and below knee amputation a week later. Five of the 23 patients (22%) who responded successfully to rfviia therapy required reexploration of the mediastinum for evacuation of clots. One of these five patients was the same patient who developed clots at multiple sites. The patient who received four doses of rfviia did not develop any clots. There was no thrombosis detected in the control group patients. Comment In our study, rfviia therapy produced a significant reduction of chest tube bleeding in 96% of pediatric cardiac surgery patients in the immediate postoperative period. One patient who failed to respond to four doses of rfviia had a surgical source of bleeding to explain the failure of therapy. The reduction in chest tube bleeding was concomitantly associated with a significant reduction in administration of blood products, including packed red cells, fresh frozen plasma, platelets, and cryoprecipitate. The hematology laboratory values in the study group were no different than the control group prior to administration of rfviia, making it unlikely that correction of coagulation disorder was responsible for control of bleeding. The mechanism of action of rfviia to control bleeding in postoperative cardiac surgery patients is unclear. On the basis of in vivo and in vitro studies it is likely that, in patients undergoing cardiac surgery on CPB, rfviia binds to tissue factor (TF) expressed by cells especially monocytes at the sites of vascular injury [13 16]. The VIIa-TF enzymatic complex activates factor X to Xa [6], leading to optimal thrombin generation and control of bleeding. Recombinant factor seven can also generate thrombin independently of TF on the surface of activated platelets, which accumulate at sites of vessel wall injury. However, thrombocytopenia and platelet dysfunction are common after cardiopulmonary bypass [17] and supraphysiologic doses of rfviia are necessary to form thrombin as the affinity of rfviia for activated platelets is low. It is possible that rfviia may have contributed to the generation of thrombin by this mechanism to some extent as all our patients received platelet transfusion prior to rfviia therapy. In our study, 30 to 50 g/kg per dose of rfviia was effective in controlling severe bleeding for all patients, including infants placed on ECMO. Recombinant factor seven was licensed in 1999 for management of hemophilia patients with inhibitors. The standard bolus dosing in these patients ranges from 90 to 120 g/kg given every two to three hours until an arrest of bleeding is seen [18]. This supraphysiologic or pharmacologic dosing of rfviia is based on its mechanism to control bleeding in these patients that include both a TF-dependent mechanism and the generation of factor Xa and IXa on the surface of activated platelets unrelated to TF. Recombinant factor seven has a very low affinity to activated platelets requiring high doses to achieve hemostasis in these patients. The frequency of rfviia dose is every two to three hours as the half life of rfviia is 2.9 hours in adults [19]. The optimal dosing in patients who do not have hemophilia has not been determined; doses as low as 20 g/kg have been found to be effective in patients who do not have hemophilia [20]. A dose of 30 g/kg elevates factor VII levels from twofold to tenfold of normal [6]. Similarly, doses of 25 g/kg have been found to control life- or limb-threatening bleeding episodes after surgery in patients with factor VII deficiency [21]. Since none of our patients had hemophilia, we initiated rfviia therapy at a lower dose on the postulation that rfviia achieves hemostasis in patients undergoing cardiopulmonary bypass predominantly by binding to TF expressed at the sites of vascular injury. The dose and frequency of rfviia administration after cardiac surgery operations, including patients placed on ECMO, has varied in the literature [8 12]. In a recent review of rfviia use in adult and pediatric patients after operations requiring cardiopulmonary bypass, rapid hemostasis was achieved after one dose of rfviia (mean, 57 g/kg) in 14 patients (70%). In the remaining six patients gradual hemostasis was achieved after a mean of 3.4 doses [22]. In our study, 62% of the patients responded to one dose and 29% of patients responded to two doses (mean dose, g/kg/ dose) administered at an interval of two to four hours. In the absence of surgical cause of bleeding, changes in temperature and acid-base balance must be looked into for failure of rfviia therapy [23]. We believe that a smaller dose administration seems to be beneficial in controlling severe bleeding without causing major thrombosis, as these patients do not have inherent coagulation disorder as seen in hemophilia patients. Administration of rfviia therapy caused a 50% reduc-

7 Ann Thorac Surg AGARWAL ET AL 2007;84:161 9 RECOMBINANT FACTOR SEVEN THERAPY tion in surgical reexploration of children having severe bleeding in our study as compared with the control group, although not statistically significant. These results are, however, biased as patients in the control group who responded to surgical reexploration were eliminated from the analysis. It is possible that study group patients may have benefited by a reduction in mediastinal exploration for bleeding. Only one of five study group patients (20%) was found to have a surgical cause of bleeding on mediastinal reexploration. A surgical cause of bleeding is identified in less than 50% of cases on reexploration in adults [3]. A reduction in mediastinal exploration in children by use of rfviia may be beneficial as reexploration after cardiac surgery operation in adult patients has been associated with prolonged mechanical ventilation, increased risk of renal failure, postoperative arrhythmias, infections, and mortality [3]. Two of the 24 patients who received rfviia in the study group developed major thrombosis. Both patients were on ECMO support at the time they developed complications. One patient developed clots in the ECMO circuit and another patient developed thrombosis at various bleeding sites, including the mediastinum, central venous catheter insertion site, and femoral artery cannulation site. In the first patient, it is possible that the preexistent small clot and increased area of turbulence in the circuit led to adhesion of platelets at that site. Administration of rfviia led to perpetuation of this small clot by binding with activated platelets on its surface. The second patient placed on ECMO had bleeding from multiple sites, including the central venous catheter, femoral arterial line attempt site, and the mediastinum. Although we are unable to totally explain the exaggerated response to rfviia in this patient, it is possible that rfviia bound to activated platelets as well as TF at multiple bleeding sites. Bleeding or thrombotic complications are known in patients placed on ECMO [24] and thrombotic complications have been reported in 2 to 18.7% of neonatal respiratory ECMO runs [25]. Complications have not been reported in the small number of pediatric ECMO patients treated with rfviia therapy after cardiac surgery [11, 12] although thrombosis has been reported in adult case reports that received rfviia on ECMO after lung transplant [26, 27]. A recent review reported 52% incidence of thromboembolic adverse effects after labeled and unlabeled rfviia therapy [28]. However, 38% of these events had concomitant use of hemostatic agents and no drug dose response of rfviia was reported [28]. We believe that examination of the ECMO circuit for any clots or turbulence prior to administration of rfviia would be advisable for patients placed on ECMO. Any significant elevation of ECMO circuit pressures after rfviia therapy may be an early indicator of clot formation in the ECMO circuit. We recommend careful and cautious consideration of rfviia therapy in patients with bleeding from multiple sites, especially arterial line sites, to avoid initiation of thrombosis at these sites. A careful search for clots in the mediastinum after rfviia therapy may be necessary in some patients as 22% of study group patients were found to have clots in the mediastinum. Delayed sternal closure in the majority of our patients facilitated early identification and evacuation of clots from the mediastinum. Our study has the limitations of its retrospective nature. Detection of postoperative bleeding based on chest tube blood losses and detection of thrombosis based on visual examination of the ECMO circuit and peripheral perfusion of the extremities have their own limitations. We did not assess the cost to benefit ratio of rfviia, which is a concerning factor in its administration given the high cost of rfviia. We believe that a properly constructed randomized prospective study will be beneficial to validate our results and to justify its use over surgical reexploration. The limited number of complications in our study makes it difficult to clearly delineate their pathophysiology. Weconclude that rfviia therapy administered in the dose range of 30 to 50 g/kg per dose every two to four hours seems to be effective in controlling severe bleeding in the postoperative period in pediatric cardiac surgery patients. A similar dose schedule is equally effective in controlling bleeding in patients placed on ECMO postoperatively. We believe that rfviia should not be routinely used to control bleeding, particularly in patients on ECMO, before more prospective studies are undertaken but rather should be considered as a rescue therapy for those patients with life threatening bleeding. Judicious use of rfviia is advocated in patients bleeding from multiple sites or having preexistent clots or turbulence in the ECMO circuit to eliminate major complications. This study has been supported by intradepartmental funds. No external financial support or grants were used to fund the study. References Manno CS, Hedberg KW, Kim HC, et al. Comparison of the hemostatic effects of fresh whole blood, stored whole blood, and components after open heart surgery in children. Blood 1991;77: Porte RJ, Leebeek FW. Pharmacological strategies to decrease transfusion requirements in patients undergoing surgery. Drugs 2002;62: Moulton MJ, Creswell LL, Mackey ME, Cox JL, Rosenbloom M. Reexploration for bleeding is a risk factor for adverse outcomes after cardiac operations. J Thorac Cardiovasc Surg 1996;111: Hedner U, Kisiel W. Use of human factor VIIa in the treatment of two hemophilia A patients with high-titer inhibitors. J Clin Invest 1983; Martinowitz U, Kenet G, Segal E, et al. Recombinant activated factor VII for adjunctive hemorrhage control in trauma. J Trauma 2001;51: Al Douri M, Shafi T, Al Khudairi D, et al. Effect of the administration of recombinant activated factor VII (rfviia; NovoSeven) in the management of severe uncontrolled bleeding in patients undergoing heart valve replacement surgery. Blood Coagul Fibrinolysis 2000;11(suppl 1):S Hendriks HG, Meijer K, de Wolf JT, et al. Reduced transfusion requirements by recombinant factor VIIa in orthotopic liver transplantation: a pilot study. Transplantation 2001;71:

8 168 AGARWAL ET AL Ann Thorac Surg RECOMBINANT FACTOR SEVEN THERAPY 2007;84: Razon Y, Erez E, Vidne B, et al. Recombinant factor VIIa (NovoSeven) as a hemostatic agent after surgery for congenital heart disease. Paediatr Anaesth 2005;15: Pychynska-Pokorska M, Moll JJ, Krajewski W, Jarosik P. The use of recombinant coagulation factor VIIa in uncontrolled postoperative bleeding in children undergoing cardiac surgery with cardiopulmonary bypass. Pediatr Crit Care Med 2004;5: Verrijckt A, Proulx F, Morneau S, Vobecky S. Activated recombinant factor VII for refractory bleeding during extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 2004;127: Wittenstein B, Ng C, Ravn H, Goldman A. Recombinant factor VII for severe bleeding during extracorporeal membrane oxygenation following open heart surgery. Pediatr Crit Care Med 2005;6: Dominguez TE, Mitchell M, Friess SH, et al. Use of recombinant factor VIIa for refractory hemorrhage during extracorporeal membrane oxygenation. Pediatr Crit Care Med 2005;6: Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS, Gimbrone MA Jr. Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: characterization and comparison with the actions of interleukin 1. Proc Natl Acad Sci U S A 1986;83: Philippou H, Adami A, Davidson SJ, Pepper JR, Burman JF, Lane DA. Tissue factor is rapidly elevated in plasma collected from the pericardial cavity during cardiopulmonary bypass. Thromb Haemost 2000;84: Chung JH, Gikakis N, Rao AK, Drake TA, Colman RW, Edmunds LH Jr. Pericardial blood activates the extrinsic coagulation pathway during clinical cardiopulmonary bypass. Circulation 1996;93: Andrew M, Paes B, Milner R, et al. Development of the human coagulation system in the full-term infant. Blood 1987;70: Kern FH, Morana NJ, Sears JJ, Hickey PR. Coagulation defects in neonates during cardiopulmonary bypass. Ann Thorac Surg 1992;54: Seremetis S. Dose optimization of recombinant factor VIIa in the treatment of acute bleeding in haemophilia-associated inhibitors. Blood Coagul Fibrinolysis 2003;14:S Lindley CM, Sawyer WT, Macik BG, et al. Pharmacokinetics and pharmacodynamics of recombinant factor VIIa. Clin Pharmacol Ther 1994;55: Friederich PW, Henny CP, Messelink EJ, et al. Effect of recombinant activated factor VII on perioperative blood loss in patients undergoing retropubic prostatectomy: a doubleblind placebo-controlled randomized trial. Lancet 2003;361: Scharrer I. Recombinant factor VIIa for patients with inhibitors to factor VIII or IX or factor VII deficiency. Haemophilia 1999;5: DiDomenico RJ, Massad MG, Kpodonu J, Navarro RA, Geha AS. Use of recombinant activated factor VII for bleeding following operations requiring cardiopulmonary bypass. Chest 2005;127: Meng ZH, Wolberg AS, Monroe DM III, Hoffman H. The effect of temperature and ph on the activity of factor VIIa: implications for the efficacy of high-dose factor VIIa in hypothermic and acidotic patients. J Trauma 2003;55: Muntean W. Coagulation and anticoagulation in extracorporeal membrane oxygenation. Artif Organs 1999;23: ELSO Data Registry Report: International Summary. University of Michigan, Ann Arbor, MI. July Bui JD, Despotis GD, Trulock EP, Patterson GA, Goodnough LT. Fatal thrombosis after administration of activated prothrombin complex concentrates in a patient supported by extracorporeal membrane oxygenation who had received activated recombinant factor VII. J Thorac Cardiovasc Surg 2002;124: Lucey MA, Myburgh JA. Recombinant activated factor VII for exsanguinating haemorrhage post bilateral lung transplantation for extra-corporeal lung support-dependent respiratory failure. Anaesth Intensive Care 2003;31: O Connell KA, Wood JJ, Wise RP, Lozier JN, Braun MM. Thromboembolic adverse events after use of recombinant human coagulation factor VIIa. JAMA 2006;295: INVITED COMMENTARY So far, the off-label use of recombinant activated factor VII (rfviia) in cardiac surgery has been driven by enthusiastic case reports that carry a well-recognized, positive submission and positive publication bias. There is no replacement for prospective, randomized, placebocontrolled data to guide routine clinical practice; case reports are better suited to postmarketing surveillance aimed at identifying adverse events and safety concerns. However, larger retrospective case series or case control studies are helpful both to identify potential safety concerns and pitfalls for prospective studies and to start deciphering the conflicting information surrounding the role of rfviia in cardiac surgery. The current case control study by Agarwal and colleagues [1] is a positive contribution to this literature and the strength of this study lies in the focus on adverse, thrombotic events, which are a major clinical concern when using rfviia. The possible extension of a preexisting clot in the extracorporeal membrane oxygenation circuitry (not an uncommon finding), and the thrombotic occlusion of a small caliber, injured vessel after rfviia are also sobering thoughts for rfviia use in the presence of intraaortic balloon pumps, ventricular assist devices, and small caliber vascular anastamoses. When the prospective studies are designed, the focus should be on outcome. Chest tube drainage is not an outcome, neither is the sparing of a few blood products in an otherwise massively transfused patient, if the consequence is an increase in morbid thrombosis. A benefit in terms of improved morbidity or mortality and a careful evaluation of safety needs to be demonstrated. What dose should we use? In cell-based models of thrombin generation, the response to a given dose of rfviia is dependent on the concentration of other coagulation proteins and platelets. Such considerations are important when planning a prospective study of rfviia, as lower doses may be effective after aggressive transfusion with higher or repeated doses needed when trying to avoid transfusion. Currently there is no consensus on the appropriate dose for off-label use, with reports ranging from 20 to 200 mcg/kg. As noted by O Connell and colleagues [2], patients with active bleeding are more at risk of thrombotic 2007 by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur