Impact of Early, Low-Dose Factor VIIa on Subsequent Transfusions and Length of Stay in Cardiac Surgery

      Highlights

      • Recombinant Factor VIIa (rFVIIa) has been used for cardiac surgical bleeding in an off-label manner
      • Previous studies have identified that low dose rFVIIa (<40 mcg/kg) may have less thrombotic potential than the higher approved doses for hemophilia (90 mcg/kg)
      • The optimal timing of when to administer rFVIIa during the course of cardiac surgical bleeding resuscitation is unknown
      • The current analyzed timing of administration of low dose rFVIIa in terms of number of allogeneic transfusions and found unclear benefit without increased adverse events

      Objective

      Recombinant factor VII (rFVIIa) is used to treat cardiac surgical bleeding in an off-label manner. However, optimal dosing and timing of administration to provide efficacious yet safe outcomes remain unknown.

      Design

      Retrospective, observational study.

      Setting

      Tertiary care academic center.

      Participants

      Cardiac surgical patients (N = 214) who received low-dose rFVIIa for cardiac surgical bleeding.

      Interventions

      Patients were allocated into one of three groups based on timing of rFVIIa administration during the course of bleeding resuscitation based on the number of hemostatic products given before rFVIIa administration: group one = ≤one products (n = 82); group two = two-to-four products (n = 73); and group three= ≥five products (n = 59).

      Measurements and Main Results

      Patients who received low-dose rFVIIa later in the course of bleeding resuscitation (group three) had longer intensive care unit stays (p = 0.014) and increased incidence of postoperative renal failure when compared with group one (p = 0.039). Total transfusions were lowest in patients who received rFVIIa early in the course of resuscitation (group one) (median, two [interquartile range (IQR), 1-4.75]) and highest in group three (median, 11 [IQR, 8-14]; p < 0.001). Subsequent blood product transfusions after rFVIIa administration were highest in group two (p = 0.003); however, the median for all three groups was two products. There were no differences in thrombosis, reexplorations, or mortality in any of the groups.

      Conclusions

      This study identified no differences in adverse outcomes based on timing of administration of low-dose rFVIIa for cardiac surgical bleeding defined by stage of resuscitation, but the benefits of early administration remain unclear.

      Graphical abstract

      Key Words

      OFF-LABEL USE of recombinant activated factor VII (rFVIIa; NovoSeven, Novo Nordisk, Bagsvaerd, Denmark) for perioperative cardiac surgical bleeding has been shown to reduce bleeding,
      • Habib AM
      • Mousa AY
      • Al-Halees Z
      Recombinant activated factor VII for uncontrolled bleeding postcardiac surgery.
      • Mayo A
      • Misgav M
      • Kluger Y
      • et al.
      Recombinant activated factor VII (NovoSeven): Addition to replacement therapy in acute, uncontrolled and life-threatening bleeding.
      • Tritapepe L
      • De Santis V
      • Vitale D
      • et al.
      Recombinant activated factor VII for refractory bleeding after acute aortic dissection surgery: A propensity score analysis.
      • Andersen ND
      • Bhattacharya SD
      • Williams JB
      • et al.
      Intraoperative use of low-dose recombinant activated factor VII during thoracic aortic operations.
      blood product administration,
      • Habib AM
      • Mousa AY
      • Al-Halees Z
      Recombinant activated factor VII for uncontrolled bleeding postcardiac surgery.
      ,
      • Gelsomino S
      • Lorusso R
      • Romagnoli S
      • et al.
      Treatment of refractory bleeding after cardiac operations with low-dose recombinant activated factor VII (NovoSeven): A propensity score analysis.
      ,
      • Romagnoli S
      • Bevilacqua S
      • Gelsomino S
      • et al.
      Small-dose recombinant activated factor VII (NovoSeven) in cardiac surgery.
      and the rate of reoperations.
      • Andersen ND
      • Bhattacharya SD
      • Williams JB
      • et al.
      Intraoperative use of low-dose recombinant activated factor VII during thoracic aortic operations.
      ,
      • Romagnoli S
      • Bevilacqua S
      • Gelsomino S
      • et al.
      Small-dose recombinant activated factor VII (NovoSeven) in cardiac surgery.
      ,
      • Gill R
      • Herbertson M
      • Vuylsteke A
      • et al.
      Safety and efficacy of recombinant activated factor VII: A randomized placebo-controlled trial in the setting of bleeding after cardiac surgery.
      However, there also has been an association with increased mortality,
      • Alfirevic A
      • Duncan A
      • You J
      • et al.
      Recombinant factor VII is associated with worse survival in complex cardiac surgical patients.
      thrombosis,
      • Bowman LJ
      • Uber WE
      • Stroud MR
      • et al.
      Use of recombinant activated factor VII concentrate to control postoperative hemorrhage in complex cardiovascular surgery.
      • O'Connell KA
      • Wood JJ
      • Wise RP
      • et al.
      Thromboembolic adverse events after use of recombinant human coagulation factor VIIa.
      • Ponschab M
      • Landoni G
      • Biondi-Zoccai G
      • et al.
      Recombinant activated factor VII increases stroke in cardiac surgery: A meta-analysis.
      stroke,
      • Gill R
      • Herbertson M
      • Vuylsteke A
      • et al.
      Safety and efficacy of recombinant activated factor VII: A randomized placebo-controlled trial in the setting of bleeding after cardiac surgery.
      ,
      • Ponschab M
      • Landoni G
      • Biondi-Zoccai G
      • et al.
      Recombinant activated factor VII increases stroke in cardiac surgery: A meta-analysis.
      ,
      • Zangrillo A
      • Mizzi A
      • Biondi-Zoccai G
      • et al.
      Recombinant activated factor VII in cardiac surgery: A meta-analysis.
      and renal morbidity.
      • Alfirevic A
      • Duncan A
      • You J
      • et al.
      Recombinant factor VII is associated with worse survival in complex cardiac surgical patients.
      ,
      • Bowman LJ
      • Uber WE
      • Stroud MR
      • et al.
      Use of recombinant activated factor VII concentrate to control postoperative hemorrhage in complex cardiovascular surgery.
      These inconsistent outcomes may be a function of variations in both timing and dosing of rFVIIa.
      Previously reported doses of rFVIIa for cardiac surgical bleeding range from 12-
      • Brase J
      • Finger B
      • He J
      • et al.
      Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
      to-90
      • Brase J
      • Finger B
      • He J
      • et al.
      Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
      ,
      • von Heymann C
      • Redlich U
      • Jain U
      • et al.
      Recombinant activated factor VII for refractory bleeding after cardiac surgery—a retrospective analysis of safety and efficacy.
      µg/kg, with <40 µg/kg considered low-dose rFVIIa. Additionally and perhaps of equal import, previous analyses report administration of rFVIIa late in bleeding resuscitation as “rescue therapy” or given for “refractory bleeding,”
      • Romagnoli S
      • Bevilacqua S
      • Gelsomino S
      • et al.
      Small-dose recombinant activated factor VII (NovoSeven) in cardiac surgery.
      ,
      • Hoffmann T
      • Assmann A
      • Dierksen A
      • et al.
      A role for very low-dose recombinant activated factor VII in refractory bleeding after cardiac surgery: Lessons from an observational study.
      ,
      • Karkouti K
      • Beattie WS
      • Wijeysundera DN
      • et al.
      Recombinant factor VIIa for intractable blood loss after cardiac surgery: A propensity score-matched case-control analysis.
      which is in line with practice advisories recommending administration of rFVIIa for severe, intractable coagulopathic bleeding
      • Raphael J
      • Mazer CD
      • Subramani S
      • et al.
      Society of Cardiovascular Anesthesiologists Clinical Practice Improvement Advisory for Management of Perioperative Bleeding and Hemostasis in Cardiac Surgery Patients.
      • Goodnough LT
      • Lublin DM
      • Zhang L
      • et al.
      Transfusion medicine service policies for recombinant factor VIIa administration.
      • Boer C
      • Meesters MI
      • Milojevic M
      • et al.
      2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery.
      or only warranted for life-threatening hemorrhage.
      • Roberts HR
      • Monroe DM
      • White GC.
      The use of recombinant factor VIIa in the treatment of bleeding disorders.
      In fact, early administration of rFVIIa for cardiac surgical bleeding rarely is reported, with advisory statements repudiating this use.
      • Brase J
      • Finger B
      • He J
      • et al.
      Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
      ,
      • Diprose P
      • Herbertson MJ
      • O'Shaughnessy D
      • et al.
      Activated recombinant factor VII after cardiopulmonary bypass reduces allogeneic transfusion in complex non-coronary cardiac surgery: Randomized double-blind placebo-controlled pilot study.
      ,
      • Boer C
      • Meesters MI
      • Milojevic M
      • et al.
      2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery.
      ,
      • Payani N
      • Foroughi M
      • Dabbagh A.
      The effect of intravenous administration of active recombinant factor VII on postoperative bleeding in cardiac valve reoperations; a randomized clinical trial.
      ,
      • Singh SP
      • Chauhan S
      • Choudhury M
      • et al.
      Recombinant activated factor VII in cardiac surgery: Single-center experience.
      Currently, the optimal timing of rFVIIa administration during the course of bleeding resuscitation in the cardiac surgical perioperative period is unknown. This study sought to evaluate if intensive care unit (ICU) length of stay and need for subsequent allogeneic transfusions were different in patients who received low-dose rFVIIa early in the course of bleeding resuscitation compared with patients who received low-dose rFVIIa later in the course of resuscitation.

      Methods

      This retrospective study was reviewed and approved by the University of Kansas Medical Center institutional review board. A total of 2,210 patients underwent cardiac surgery requiring cardiopulmonary bypass (CPB) between January 2014 and January 2018. Of these patients, 214 patients (9.7%) received low-dose rFVIIa perioperatively for cardiac surgical bleeding based on an institutional guideline for rFVIIa administration (Fig 1) (Supplement 1).
      Fig 1
      Fig 1Administration algorithm of low-dose rFVIIa based on an insititutional guideline for bleeding cardiac surgical patients. rFVIIa, recombinant factor VII.
      Figure 2 depicts patient allocation into one of three groups based on stage of bleeding resuscitation at the time of rFVIIa administration. There were no guidelines concerning administration of allogeneic transfusions, and these products were administered at the discretion of the perioperative providers. The groups were defined as early, or group one, if they received ≤one unit of a hemostatic component before rFVIIa administration (n = 82); in the middle of resuscitation, or group two, if they received two-to-four units of any composition of hemostatic components before rFVIIa administration (n = 73); and late in the course of resuscitation, or group three, if ≥five units were given before rFVIIa administration (group three) (n = 59). Hemostatic components were defined as pooled platelets, fresh frozen plasma (FFP), or cryoprecipitate. Packed red blood cells were not included in defining the stage of resuscitation due to lack of hemostatic effects. The definitions of stage of bleeding resuscitation were derived from previously published definitions of massive transfusion,
      • Savage SA
      • Sumislawski JJ
      • Zarzaur BL
      • et al.
      The new metric to define large-volume hemorrhage: Results of a prospective study of the critical administration threshold.
      along with investigator consensus from experience with cardiac surgical bleeding.
      Fig 2
      Fig 2Selection pathway for patients included in this retrospective, observational study of low-dose rFVIIa administration for cardiac surgical bleeding. rFVIIa, recombinant factor VII.
      All patients received prophylactic aminocaproic acid (Amicar, Clover Pharmaceuticals Corp, Chengdu, China) with an intravenous bolus of 5 g before initiation of bypass, followed by an infusion of 2 g/h for 6.25 hours. If the procedure was still ongoing after 6.25 hours, an additional 12.5- g infusion of aminocaproic acid was given. Heparin was administered before bypass as determined by the Hepcon HMS system (Medtronic, Minneapolis, MN), with most doses consisting of 300 units/kg of intravenous heparin bolus, for a goal activated clotting time of 480 seconds, followed by calculated intravenous protamine reversal ranging from 200-to-250 mg. No patients received prothrombin complex concentrate or vitamin K.
      Primary outcomes were number of products given after low-dose rFVIIa administration and ICU length of stay. Total products given after rFVIIa administration were calculated as the sum of each unit of hemostatic components and packed red blood cell units. Secondary outcomes included individual blood product components, mechanical ventilation time, 30-day mortality, renal failure, infection (sepsis, pneumonia, surgical site infection, and composites of these), thrombosis (stroke, transient ischemia attack, pulmonary embolism, deep vein thrombosis, and composites of these), and need for reoperation for bleeding and cardiac arrest. Postoperative renal failure was defined as a creatinine increase of 300% above preoperative value or a creatinine >4 mg/dL with at least a 0.5 mg/dL rise.
      • Bellomo R
      • Ronco C
      • Kellum JA
      • et al.
      Acute renal failure – definition, outcome measures, animal models, fluid therapy and information technology needs: The Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group.

       Statistical Analysis

      Baseline characteristics of patients from the three groups were summarized using medians and interquartile ranges (IQRs) for continuous variables and percentages for categorical variables. For univariate analyses, Fisher's exact test or Pearson chi-squared test was used to examine associations between categorical variables and the rFVIIa groups. Nonparametric tests were used for associations between continuous variables and rFVIIa groups.
      To study group differences in the primary and secondary outcomes adjusting for covariates, linear regressions were conducted for continuous outcomes. Due to similar clinical risks for each outcome, a full model, including the group variable and all other confounding variables, was used with a subsequent backward selection with a p value threshold of 0.1, while forcing the group variable in the model. After backward selections, the significance of the group variable was obtained by the final model adjusting of remaining confounding variables. Logistic regressions were conducted for binary outcomes and Poisson regressions for count outcomes, such as blood product usage, which were not treated as Gaussian outcomes. Kruskal-Wallis rank sum test was used for continuous variables, whereas categorical variables were analyzed using Pearson chi-squared test or Fisher's exact test if the outcome was sparse. A transformation was done for a continuous outcome if needed for normality assumption. The authors considered baseline characteristics including dose of rFVIIa, age, sex, body mass index, surgical procedure (aortic v non-aortic), race (white v nonwhite), emergent or urgent surgery, CPB time, preoperative anticoagulants within 48 hours, aspirin within five days of surgery, preoperative ejection fraction, chronic lung disease, diabetes, and tobacco use. All statistical analyses were conducted using R software (The R Foundation for Statistical Computing, Vienna, Austria).

      Results

      Overall, 214 patients were administered low-dose rFVIIa, with a median dose of 13.5 µg/kg. Baseline characteristics and preoperative characteristics are summarized and compared across the three groups in Table 1. The doses of rFVIIa were significantly different among the three groups, with patients in group three, who received rFVIIa later in the course of resuscitation, receiving higher doses of rFVIIa than those in groups one and two (18.02 µg/kg v 12.16 µg/kg and 14.08 µg/kg, respectively, p < 0.001). Group three patients were more often female (p = 0.004), were without diabetes mellitus (p < 0.001), underwent aortic surgeries (p < 0.001), were more often emergent or urgent (p < 0.001), and had longer CPB times (p < 0.001).
      Table 1Baseline Characteristics of Patients Who Received Low-Dose rFVIIa in Three Preidentified Coagulation Product Transfusion Groups
      Total (N = 214)Group 1 (n = 82):

      0-1 Products
      Group 2 (n = 73):

      2-4 Products
      Group 3 (n = 59):

      ≥5 Products
      p Value
      p values were obtained using Kruskal-Wallis rank sum test for continuous variables; Fisher's exact test or Pearson chi-squared test for categorical variables.
      rFVIIa dose (µg/kg)
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      12.16 (9.74-16.42)14.08 (10.50-23.39)18.02 (11.78-28.95)< 0.001
      Total rFVIIa dose (mg)
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      1 (1-1.5)1 (1-2)1.5 (1-2)0.006
      Pre-rFVIIa FFP units
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      0 (0-0)2 (0-2)6 (4-6)< 0.001
      Pre-rFVIIa platelet units
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      1 (0-1)1 (0-2)2 (1-2)< 0.001
      Pre-rFVIIa cryoprecipitate units
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      0 (0-0)0 (0-0)0 (0-0.5)< 0.001
      Age (y)
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      67 (54.5-73)63 (57-72)61 (48.5-71)0.195
      Male sex (n = 214)70 (85.37%)52 (71.23%)36 (61.02%)0.004
      BMI
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 214)
      28.48 (24.21-31.89)27.82 (25.05-31.40)27.76 (24.58-30.98)0.563
      Procedure (n = 214)< 0.001
       Isolated CABG35 (42.68)17 (23.29)3 (5.08)
       Valvular27 (32.93)26 (35.62)12 (20.34)
       Aortic15 (18.29)24 (32.88)41 (69.49)
       Other5 (6.10)6 (8.22)3 (5.08)
      Race (n = 210)(1 missing)(3 missing)0.790
       White71 (87.65)58 (82.86)52 (88.14)
       Black6 (7.41)5 (7.14)3 (5.08)
       Other4 (4.94)7 (10.00)4 (6.78)
      Emergent/urgent procedure (n = 214)31 (37.80)30 (41.10)41 (69.49)< 0.001
      Cardiopulmonary bypass time (min)
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 209)
      111 (87.25-158.75)

      (4 missing)
      134 (99-183)183.5 (146-228.5)

      (1 missing)
      < 0.001
      Preoperative anticoagulants (n = 213)16 (19.51)20 (27.40)15 (25.86)

      (1 missing)
      0.477
      Aspirin within 5 days of surgery (n = 211)62 (75.61)52 (73.24)

      (2 missing)
      35 (60.34)

      (1 missing)
      0.124
      Cardiogenic shock (n = 211)2 (2.50)

      (2 missing)
      1 (1.39)

      (1 missing)
      3 (5.08)0.513
      Preoperative ejection fraction %
      Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      (n = 196)
      57.5 (45-60)

      (6 missing)
      55 (43.75-65)

      (5 missing)
      55 (40-60)

      (7 missing)
      0.946
      ESRD (n = 214)4 (4.88)6 (8.22)1 (1.69)0.241
      Chronic lung disease (n = 213)25 (30.49)18 (24.66)9 (15.52)

      (1 missing)
      0.127
      Diabetes (n = 213)26 (31.71)27 (36.99)5 (8.62)

      (1 missing)
      < 0.001
      Tobacco use (n = 213)52 (63.41)35 (47.95)27 (46.55)

      (1 missing)
      0.072
      NOTE. Data presented as n (%) or as otherwise indicated. Groups were divided based on number of coagulation products (platelet, FFP, cryoprecipitate) given before rFVIIa administration for cardiac surgical bleeding. Preoperative anticoagulants defined as heparin drip or LMWH within 48 hours of surgery or adenosine diphosphate receptor inhibitors, glycoprotein IIb/IIIa inhibitors, factor Xa inhibitors, thrombolytics, or warfarin within five days of surgery.
      Abbreviations: BMI, body mass index; CABG, coronary artery bypass graft; ESRD, end-stage renal disease; FFP, fresh frozen plasma; IQR, interquartile range; LMWH, low molecular weight heparin; rFVIIa, recombinant factor VII.
      low asterisk Data presented as median (IQR, 1-3) with numbers of non-missing in first column.
      p values were obtained using Kruskal-Wallis rank sum test for continuous variables; Fisher's exact test or Pearson chi-squared test for categorical variables.
      For the primary outcome of total subsequent products after rFVIIa administration, when compared to group one, patients in group two received more total blood products after rFVIIa administration (p = 0.003), though a median of two units was given to all three groups (Table 2). The difference largely was due to significantly more FFP administration to patients in group two (p = 0.001).
      Table 2Primary and Secondary Outcomes of Patients Who Received Low-Dose rFVIIa in Three Preidentified Coagulation Product Transfusion Groups Based on Number of Coagulation Products (Platelet, Fresh Frozen Plasma (FFP), Cryoprecipitate) Given Before rFVIIa Administration for Cardiac Surgical Bleeding
      Total (n = 214)Group 1 (n = 82):

      0-1 Products
      Group 2 (n = 73):

      2-4 Products
      Group 3 (n = 59):

      ≥5 Products
      p Value
      p values were obtained from regression models adjusting for covariates.
      p Value# Group 2 v 1p Value# Group 3 v 1
      Total blood products received after rFVIIa administration
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 189)
      2 (0-4)2 (1-4)2 (1-7)< 0.0010.0030.133
       PRBC units after rFVIIa
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 195)
      0.5 (0-2)1 (0-2)1 (0-2)0.1830.0770.209
       FFP units after rFVIIa
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 190)
      0 (0-0.75)0 (0-2)0 (0-1.5)< 0.0010.0010.038
       Platelet units after rFVIIa
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 214)
      0 (0-1)1 (0-1)1 (0-1)0.8070.7620.740
       Cryoprecipitate units after rFVIIa
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 194)
      0 (0-0)0 (0-2)0 (0-2)0.4090.2500.257
      ICU length of stay (h)
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      68.83 (47.50-97.03)

      (2 missing)
      70.60 (48.00-119.50)120.00 (51.20-216.81)

      (1 missing)
      ICU length of stay (log 10 h)
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 211)
      1.84 (1.68-1.99)

      (2 missing)
      1.85 (1.68-2.08)2.08 (1.71-2.34)

      (1 missing)
      0.0420.4220.014
      Postoperative ventilation hours
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      15.12 (9.00-21.00) (5 missing)16.00 (10.00-28.05)22.71 (13.12-87.00)

      (1 missing)
      Postoperative ventilation hours (log10 h)
      Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      (n = 188)
      1.18 (0.95-1.32)

      (5 missing)
      1.20 (1.00-1.45)1.36 (1.12-1.94)

      (1 missing)
      0.601

      0.7720.323
      30-day mortality (n = 190)7 (8.54)10 (13.708 (13.56)0.2970.6290.360
      Postoperative renal failure (n = 194)2 (2.67)

      (7 missing)
      6 (8.96)

      (6 missing)
      11 (19.64)

      (3 missing)
      0.0780.1500.039
      Postoperative composite thrombosis (n = 214)4 (4.88)2 (2.74)3 (5.08)0.5610.3270.406
       Postoperative DVT (n = 214)1 (1.22)0 (0.00)1 (1.69)0.737
      p values were obtained from Fisher's exact test due to sparsity.
      1.000
      p values were obtained from Fisher's exact test due to sparsity.
      1.000
      p values were obtained from Fisher's exact test due to sparsity.
       Postoperative PE (n = 214)0 (0.00)0 (0.00)1 (1.69)0.276
      p values were obtained from Fisher's exact test due to sparsity.
      1.000
      p values were obtained from Fisher's exact test due to sparsity.
      0.418
      p values were obtained from Fisher's exact test due to sparsity.
       Postoperative stroke or TIA (n = 214)3 (3.66)2 (2.74)2 (3.39)0.4210.4000.194
      Postoperative composite infection (n = 208)5 (6.10)4 (5.48)9 (15.25)0.3650.7710.291
       Postoperative sepsis (n = 213)1 (1.22)1 (1.37)1 (1.69)0.6420.6300.381
       Postoperative pneumonia (n = 188)2 (2.44)3 (4.11)8 (13.56)0.1740.6160.094
       Postoperative surgical site infection (n = 214)2 (2.44)0 (0.00)1 (1.69)0.496
      p values were obtained from Fisher's exact test due to sparsity.
      0.498
      p values were obtained from Fisher's exact test due to sparsity.
      1.000
      p values were obtained from Fisher's exact test due to sparsity.
      Postoperative reoperation (n = 213)15 (18.29)17 (23.29)12 (20.34)0.8180.5270.799
      Postoperative cardiac arrest (n = 210)1 (1.22)8 (10.96)0 (0.00)0.002
      p values were obtained from Fisher's exact test due to sparsity.
      0.013
      p values were obtained from Fisher's exact test due to sparsity.
      1.000
      p values were obtained from Fisher's exact test due to sparsity.
      NOTE. Data presented as n (%) or as otherwise indicated.
      Abbreviations: DVT, deep venous thrombosis; ICU, intensive care unit; IQR, interquartile range; PE, pulmonary embolus; PRBC, packed red blood cell; rFVIIa, recombinant factor VII; TIA, transient ischemic attack.
      low asterisk Data presented in median (IQR, 1-3) with numbers of samples used in regression models in the first column.
      p values were obtained from regression models adjusting for covariates.
      p values were obtained from Fisher's exact test due to sparsity.
      For the primary outcome of ICU hours, compared with group one, group three patients had longer ICU hours (120.00 [IQR, 51.20-216.81] v 68.83 [IQR, 47.50-97.03], p = 0.014), with no significant differences between groups one and two. For secondary outcomes, group three had an increased incidence of renal failure compared with group one (19.64% v 2.67%, p = 0.039). There were no differences for 30-day mortality, postoperative infection, thrombosis, or reoperation for bleeding among the three groups after adjusting for relevant covariates.

      Discussion

      Excessive bleeding occurs in 3%-to-21% of patients undergoing cardiac surgery,
      • Vuylsteke A
      • Pagel C
      • Gerrard C
      • et al.
      The Papworth Bleeding Risk Score: A stratification scheme for identifying cardiac surgery patients at risk of excessive early postoperative bleeding.
      with a surgical bleeding site detected in less than half of reexplored patients.
      • Romagnoli S
      • Bevilacqua S
      • Gelsomino S
      • et al.
      Small-dose recombinant activated factor VII (NovoSeven) in cardiac surgery.
      Although blood product administration for cardiac surgical bleeding may be necessary, it is associated with increases in morbidity and mortality in a dose-dependent fashion.
      • Koch CG
      • Li L
      • Duncan AI
      • et al.
      Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting.
      Thus, the authors sought to identify whether early administration of very-low-dose rFVIIa (median, 13.5 µg/kg) in the course of bleeding resuscitation, as defined by the number of transfused products, would decrease subsequent transfusions and the associated risks. To do this, all patients who received rFVIIa for cardiac surgical bleeding were allocated into one of three groups (early, mid, and late) based on the number of transfusions given before rFVIIa administration.
      Patients in group three who received rFVIIa later in the course of bleeding resuscitation had significantly longer ICU length of stay (p = 0.014) (Fig 3) and incidence of renal failure (p = 0.039) compared with patients who received rFVIIa early in the course of hemostatic product administration (group one). These two outcomes may cohere because patients who develop any stage of acute kidney injury after cardiac surgery have been shown to have mean ICU lengths of stay of six days compared with two days for those patients who did not develop acute kidney injury.
      • Hobson C
      • Ozrazgat-Baslanti T
      • Kuxhausen A
      • et al.
      Cost and mortality associated with postoperative acute kidney injury.
      Numerous studies have demonstrated that allogeneic blood transfusions in cardiac surgery are associated with a risk-adjusted increase in postoperative renal failure.
      • Koch CG
      • Li L
      • Duncan AI
      • et al.
      Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting.
      ,
      • Karkouti K.
      Transfusion and risk of acute kidney injury in cardiac surgery.
      Thus, any decrease in allogeneic transfusions is meaningful in improving renal outcomes in a dose-dependent manner.
      • Koch CG
      • Li L
      • Duncan AI
      • et al.
      Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting.
      Alternatively, increased ICU length of stay also can be attributed to postoperative bleeding.
      • Al-Attar N
      • Johnston S
      • Jamous N
      • et al.
      Impact of bleeding complications on length of stay and critical care utilization in cardiac surgery patients in England.
      Fig 3
      Fig 3Comparison of ICU length of stay among three groups of cardiac surgical patients who received low-dose rFVIIa for bleeding at various stages of bleeding resuscitation. Group one (n = 82) received rFVIIa as first-line hemostatic therapy or after one hemostatic product (fresh frozen plasma [FFP], platelet, or cyroprecipiate) had been given. Group two (n = 73) received rFVIIa after two-to-four hemostatic products. Group three (n = 59) received rFVIIa later in the course of bleeding resuscitation after five or more hemostatic products had been given. ICU, intensive care unit; rFVIIa, recombinant factor VII.
      The total number of overall transfusions when considering pre- and post-rFVIIa administration was significantly different among the three groups (p < 0.001) and lowest in patients who received rFVIIa as first- or second-line therapy (group one) (IQR, 2 [1-4.75]), and highest in group three (IQR, 11 [8-14]). Overall, after receiving rFVIIa, all patients received a median of two subsequent allogeneic transfusions.
      Patients receiving rFVIIa in mid-stage resuscitation (group two) had higher total need for subsequent allogeneic transfusions than group one (p = 0.003) due heavily to additional FFP (p = 0.001) (Table 2; Fig 4). Interestingly, patients in groups one and three did not have a significant difference in the number of subsequent transfusions. This may be because, by definition, patients in group three already had received a higher number of transfusions before rFVIIa administration.
      Fig 4
      Fig 4Number of subsequent blood products (packed red blood cell [PRBC], fresh frozen plasma [FFP], platelets, or cyroprecipate) administered after low-dose rFVIIa was given for cardiac surgical bleeding. Group one (n = 82) received rFVIIa as first-line hemostatic therapy or after one hemostatic product (FFP, platelet, or cyroprecipiate) had been given. Group two (n = 73) received rFVIIa after two-to-four hemostatic products. Group three (n = 59) received rFVIIa later in the course of bleeding resuscitation after five or more hemostatic products had been given. rFVIIa, recombinant factor VII.
      Notably, baseline characteristics varied among the three groups, with patients in group three (≥five products received before rFVIIa administration and larger dose of rFVIIa) having more female patients, more aortic surgeries, longer CPB times, and more urgent or emergent procedures (Table 1). Each of these factors previously has been associated with increased risk of cardiac surgical bleeding.
      • Vuylsteke A
      • Pagel C
      • Gerrard C
      • et al.
      The Papworth Bleeding Risk Score: A stratification scheme for identifying cardiac surgery patients at risk of excessive early postoperative bleeding.
      ,
      • Jimenez Rivera JJ
      • Iribarren JL
      • Raya JM
      • et al.
      Factors associated with excessive bleeding in cardiopulmonary bypass patients: A nested case-control study.
      This also is supported by the findings that group three patients required more blood transfusions and higher doses of rFVIIa. Thus, associations to outcomes concerning rFVIIa in this group are less robust due to potential confounding risks.
      Although ICU length of stay was longest in group three, there were no differences among any of the groups in mechanical ventilation time. This lack of differences in ventilation hours likely was due to adherence to a strictly followed early extubation protocol applied to all cardiac surgical ICU patients, even those deemed to be high risk, at the study institution.
      • Flynn BC
      • He J
      • Richey M
      • et al.
      Early extubation without increased adverse events in high-risk cardiac surgical patients.
      ,
      • Richey M
      • Mann A
      • He J
      • et al.
      Implementation of an early extubation protocol in cardiac surgical patients decreased ventilator time but not intensive care unit or hospital length of stay.
      There were no differences in adverse events in thrombosis, infections, reexplorations for bleeding, or cardiac arrests. In theory, thrombotic events could be less with rFVIIa versus other hemostatic agents based on the mechanism of action of rFVIIa, which is shown to be localized to the site of vascular injury where tissue factor is exposed.
      • Levy JH
      • Fingerhut A
      • Brott T
      • et al.
      Recombinant factor VIIa in patients with coagulopathy secondary to anitcoagulant therapy, cirrhosis, or severe traumatic injury: Review of safety profile.
      Previous reports have shown benefits when rFVIIa is used for cardiac surgical bleeding
      • Tritapepe L
      • De Santis V
      • Vitale D
      • et al.
      Recombinant activated factor VII for refractory bleeding after acute aortic dissection surgery: A propensity score analysis.
      ,
      • Andersen ND
      • Bhattacharya SD
      • Williams JB
      • et al.
      Intraoperative use of low-dose recombinant activated factor VII during thoracic aortic operations.
      ,
      • Gill R
      • Herbertson M
      • Vuylsteke A
      • et al.
      Safety and efficacy of recombinant activated factor VII: A randomized placebo-controlled trial in the setting of bleeding after cardiac surgery.
      ,
      • Brase J
      • Finger B
      • He J
      • et al.
      Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
      ; however, disparities in outcomes remain. These disparities likely are a function of the timing and dosing strategies used when administering rFVIIa.
      • Alfirevic A
      • Duncan A
      • You J
      • et al.
      Recombinant factor VII is associated with worse survival in complex cardiac surgical patients.
      ,
      • Bowman LJ
      • Uber WE
      • Stroud MR
      • et al.
      Use of recombinant activated factor VII concentrate to control postoperative hemorrhage in complex cardiovascular surgery.
      ,
      • Ponschab M
      • Landoni G
      • Biondi-Zoccai G
      • et al.
      Recombinant activated factor VII increases stroke in cardiac surgery: A meta-analysis.
      Later administration of rFVIIa may be due to practice recommendations advising the use of rFVIIa for cardiac surgical bleeding as “rescue therapy” in times of “refractory hemorrhage” after numerous other products have been administered,
      • Raphael J
      • Mazer CD
      • Subramani S
      • et al.
      Society of Cardiovascular Anesthesiologists Clinical Practice Improvement Advisory for Management of Perioperative Bleeding and Hemostasis in Cardiac Surgery Patients.
      ,
      • Goodnough LT
      • Lublin DM
      • Zhang L
      • et al.
      Transfusion medicine service policies for recombinant factor VIIa administration.
      or warranted only as a “life-saving therapy.”
      • Boer C
      • Meesters MI
      • Milojevic M
      • et al.
      2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery.
      ,
      • Roberts HR
      • Monroe DM
      • White GC.
      The use of recombinant factor VIIa in the treatment of bleeding disorders.
      However, the risk of mortality due to uncontrolled bleeding is significant.
      • Levy JH
      • Fingerhut A
      • Brott T
      • et al.
      Recombinant factor VIIa in patients with coagulopathy secondary to anitcoagulant therapy, cirrhosis, or severe traumatic injury: Review of safety profile.
      Indeed, early-administration rFVIIa has demonstrated favorable outcomes that are not reflected in these recommendations.
      • Brase J
      • Finger B
      • He J
      • et al.
      Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
      ,
      • Payani N
      • Foroughi M
      • Dabbagh A.
      The effect of intravenous administration of active recombinant factor VII on postoperative bleeding in cardiac valve reoperations; a randomized clinical trial.
      Additionally, the low-dose strategy of rFVIIa administration likely imparts less thrombotic risk compared with many previous reports strategies of >80 µg/kg dosing.
      • Levi M
      • Levy JH
      • Andersen HF
      • et al.
      Safety of recombinant activated factor VII in randomized clinical trials.
      Costs associated with hemostatic therapies are not negligible and should be considered when treating cardiac surgical bleeding. Although this study did not conduct a therapeutic cost analysis, Table 3 demonstrates typical academic charges to patients for agents and products administered for cardiac surgical bleeding.
      Table 3Approximate Costs to a Patient in US Dollars for Administration of Therapies for Cardiac Surgical Bleeding
      TherapyCost to Patient
      Red blood cells$989
      Platelets$2054
      Fresh frozen plasma$208
      Cryoprecipitate (pool of 5 units)$1320
      rFVIIa (1 mg)$17,000
      Prothrombin complex concentrate (1,000 units)$18,000
      Abbreviation: rFVIIa, recombinant factor VII.

       Strengths and Limitations

      Although guided by an institutional policy for administration of rFVIIa, due to the retrospective nature of this study, there was potential for selection bias. Similarly, although replenishment of fundamental coagulation markers is necessary and was attempted per the institutional rFVIIa administration policy, confirmatory laboratory results were not included in this data set. Other tests of coagulation, such as thromboelastography or thromboelastometry, were not part of the rFVIIa administration pathway used in this study due to the desire for early rFVIIa administration, which may differ from other institutions. Although such practice may not be generalizable to all institutions, it reflects a single-center clinical practice and demonstrates no adverse outcomes. Additionally, the large number of secondary outcomes may lead to type-one error and are intended to be exploratory findings requiring further study.

      Conclusions

      This study sought to identify whether early, low-dose rFVIIa administration had the potential to decrease allogeneic blood product transfusions and ICU length of stay. Although broad conclusions concerning these outcomes are difficult to ascertain due to inherent limitations, there were no differences in adverse outcomes related to timing of rFVIIa administration. Thus, future randomized controlled trials may establish the optimal timing and dosing strategy for low-dose rFVIIa for cardiac surgical bleeding.

      Conflict of Interest

      The authors have no disclosures or conflicts of interest related to this study.

      References

        • Habib AM
        • Mousa AY
        • Al-Halees Z
        Recombinant activated factor VII for uncontrolled bleeding postcardiac surgery.
        J Saudi Heart Assoc. 2016; 28: 222-231
        • Mayo A
        • Misgav M
        • Kluger Y
        • et al.
        Recombinant activated factor VII (NovoSeven): Addition to replacement therapy in acute, uncontrolled and life-threatening bleeding.
        Vox Sang. 2004; 87: 34-40
        • Tritapepe L
        • De Santis V
        • Vitale D
        • et al.
        Recombinant activated factor VII for refractory bleeding after acute aortic dissection surgery: A propensity score analysis.
        Crit Care Med. 2007; 35: 1685-1690
        • Andersen ND
        • Bhattacharya SD
        • Williams JB
        • et al.
        Intraoperative use of low-dose recombinant activated factor VII during thoracic aortic operations.
        Ann Thorac Surg. 2012; 93: 1921-1928
        • Gelsomino S
        • Lorusso R
        • Romagnoli S
        • et al.
        Treatment of refractory bleeding after cardiac operations with low-dose recombinant activated factor VII (NovoSeven): A propensity score analysis.
        Eur J Cardiothorac Surg. 2008; 33: 64-71
        • Romagnoli S
        • Bevilacqua S
        • Gelsomino S
        • et al.
        Small-dose recombinant activated factor VII (NovoSeven) in cardiac surgery.
        Anesth Analg. 2006; 102: 1320-1326
        • Gill R
        • Herbertson M
        • Vuylsteke A
        • et al.
        Safety and efficacy of recombinant activated factor VII: A randomized placebo-controlled trial in the setting of bleeding after cardiac surgery.
        Circulation. 2009; 120: 21-27
        • Alfirevic A
        • Duncan A
        • You J
        • et al.
        Recombinant factor VII is associated with worse survival in complex cardiac surgical patients.
        Ann Thorac Surg. 2014; 98: 618-624
        • Bowman LJ
        • Uber WE
        • Stroud MR
        • et al.
        Use of recombinant activated factor VII concentrate to control postoperative hemorrhage in complex cardiovascular surgery.
        Ann Thorac Surg. 2008; 85: 1669-1676
        • O'Connell KA
        • Wood JJ
        • Wise RP
        • et al.
        Thromboembolic adverse events after use of recombinant human coagulation factor VIIa.
        JAMA. 2006; 295: 293-298
        • Ponschab M
        • Landoni G
        • Biondi-Zoccai G
        • et al.
        Recombinant activated factor VII increases stroke in cardiac surgery: A meta-analysis.
        J Cardiothorac Vasc Anesth. 2011; 25: 804-810
        • Zangrillo A
        • Mizzi A
        • Biondi-Zoccai G
        • et al.
        Recombinant activated factor VII in cardiac surgery: A meta-analysis.
        J Cardiothorac Vasc Anesth. 2009; 23: 34-40
        • Brase J
        • Finger B
        • He J
        • et al.
        Analysis of outcomes using low-dose and early administration of recombinant activated factor VII in cardiac surgery.
        Ann Thorac Surg. 2016; 102: 35-40
        • Diprose P
        • Herbertson MJ
        • O'Shaughnessy D
        • et al.
        Activated recombinant factor VII after cardiopulmonary bypass reduces allogeneic transfusion in complex non-coronary cardiac surgery: Randomized double-blind placebo-controlled pilot study.
        Br J Anaesth. 2005; 95: 596-602
        • von Heymann C
        • Redlich U
        • Jain U
        • et al.
        Recombinant activated factor VII for refractory bleeding after cardiac surgery—a retrospective analysis of safety and efficacy.
        Crit Care Med. 2005; 33: 2241-2246
        • Hoffmann T
        • Assmann A
        • Dierksen A
        • et al.
        A role for very low-dose recombinant activated factor VII in refractory bleeding after cardiac surgery: Lessons from an observational study.
        J Thorac Cardiovasc Surg. 2018; 156: 1564-1573
        • Karkouti K
        • Beattie WS
        • Wijeysundera DN
        • et al.
        Recombinant factor VIIa for intractable blood loss after cardiac surgery: A propensity score-matched case-control analysis.
        Transfusion. 2005; 45: 26-34
        • Raphael J
        • Mazer CD
        • Subramani S
        • et al.
        Society of Cardiovascular Anesthesiologists Clinical Practice Improvement Advisory for Management of Perioperative Bleeding and Hemostasis in Cardiac Surgery Patients.
        Anesth Analg. 2019; 129: 1209-1221
        • Goodnough LT
        • Lublin DM
        • Zhang L
        • et al.
        Transfusion medicine service policies for recombinant factor VIIa administration.
        Transfusion. 2004; 44: 1325-1331
        • Boer C
        • Meesters MI
        • Milojevic M
        • et al.
        2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery.
        J Cardiothorac Vasc Anesth. 2018; 32: 88-120
        • Roberts HR
        • Monroe DM
        • White GC.
        The use of recombinant factor VIIa in the treatment of bleeding disorders.
        Blood. 2004; 104: 3858-3864
        • Payani N
        • Foroughi M
        • Dabbagh A.
        The effect of intravenous administration of active recombinant factor VII on postoperative bleeding in cardiac valve reoperations; a randomized clinical trial.
        Anesth Pain Med. 2015; 5: e22846
        • Singh SP
        • Chauhan S
        • Choudhury M
        • et al.
        Recombinant activated factor VII in cardiac surgery: Single-center experience.
        Asian Cardiovasc Thorac Ann. 2014; 22: 148-154
        • Savage SA
        • Sumislawski JJ
        • Zarzaur BL
        • et al.
        The new metric to define large-volume hemorrhage: Results of a prospective study of the critical administration threshold.
        J Trauma Acute Care Surg. 2015; 78: 224-229
        • Bellomo R
        • Ronco C
        • Kellum JA
        • et al.
        Acute renal failure – definition, outcome measures, animal models, fluid therapy and information technology needs: The Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group.
        Crit Care. 2004; 8: R204-R212
        • Vuylsteke A
        • Pagel C
        • Gerrard C
        • et al.
        The Papworth Bleeding Risk Score: A stratification scheme for identifying cardiac surgery patients at risk of excessive early postoperative bleeding.
        Eur J Cardiothorac Surg. 2011; 39: 924-930
        • Koch CG
        • Li L
        • Duncan AI
        • et al.
        Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting.
        Crit Care Med. 2006; 34: 1608-1616
        • Hobson C
        • Ozrazgat-Baslanti T
        • Kuxhausen A
        • et al.
        Cost and mortality associated with postoperative acute kidney injury.
        Ann Surg. 2015; 261: 1207-1214
        • Karkouti K.
        Transfusion and risk of acute kidney injury in cardiac surgery.
        Br J Anaesth. 2012; 109: i29-i38
        • Al-Attar N
        • Johnston S
        • Jamous N
        • et al.
        Impact of bleeding complications on length of stay and critical care utilization in cardiac surgery patients in England.
        J Cardiothorac Surg. 2019; 14: 64
        • Jimenez Rivera JJ
        • Iribarren JL
        • Raya JM
        • et al.
        Factors associated with excessive bleeding in cardiopulmonary bypass patients: A nested case-control study.
        J Cardiothorac Surg. 2007; 2: 17
        • Flynn BC
        • He J
        • Richey M
        • et al.
        Early extubation without increased adverse events in high-risk cardiac surgical patients.
        Ann Thorac Surg. 2019; 107: 453-459
        • Richey M
        • Mann A
        • He J
        • et al.
        Implementation of an early extubation protocol in cardiac surgical patients decreased ventilator time but not intensive care unit or hospital length of stay.
        J Cardiothorac Vasc Anesth. 2018; 32: 739-744
        • Levy JH
        • Fingerhut A
        • Brott T
        • et al.
        Recombinant factor VIIa in patients with coagulopathy secondary to anitcoagulant therapy, cirrhosis, or severe traumatic injury: Review of safety profile.
        Transfusion. 2006; 46: 919-933
        • Levi M
        • Levy JH
        • Andersen HF
        • et al.
        Safety of recombinant activated factor VII in randomized clinical trials.
        N Engl J Med. 2010; 363: 1791-1800