This study attempts to determine cost-effectiveness of fibrinogen concentrate vs cryoprecipitate for managing active bleeding in adult patients who underwent cardiac surgery.
Key Points
Question
Is fibrinogen concentrate (FC) cost-effective when compared with cryoprecipitate for adult patients who underwent cardiac surgery and experienced active bleeding resulting in acquired hypofibrinogenemia?
Findings
In this economic evaluation, after exclusion of patients who were critically ill in whom there was a large variability in costs, FC was found to be cost-effective in comparison with cryoprecipitate.
Meaning
Fibrinogen concentrate may be cost-effective for bleeding management in adult patients who underwent cardiac surgery and experienced bleeding resulting in acquired hyperfibrinogenemia.
Abstract
Importance
Excessive bleeding requiring fibrinogen replacement is a serious complication of cardiac surgery. However, the relative cost-effectiveness of the 2 available therapies—fibrinogen concentrate and cryoprecipitate—is unknown.
Objective
To determine cost-effectiveness of fibrinogen concentrate vs cryoprecipitate for managing active bleeding in adult patients who underwent cardiac surgery.
Design, Setting, and Participants
A within-trial economic evaluation of the Fibrinogen Replenishment in Surgery (FIBERS) randomized clinical trial (February 2017 to November 2018) that took place at 4 hospitals based in Ontario, Canada, hospitals examined all in-hospital resource utilization costs and allogeneic blood product (ABP) transfusion costs incurred within 28 days of surgery. Participants included a subset of 495 adult patients from the FIBERS trial who underwent cardiac surgery and developed active bleeding and acquired hypofibrinogenemia requiring fibrinogen replacement.
Interventions
Fibrinogen concentrate (4 g per dose) or cryoprecipitate (10 units per dose) randomized (1:1) up to 24 hours postcardiopulmonary bypass.
Main Outcomes and Measures
Effectiveness outcomes included number of ABPs administered within 24 hours and 7 days of cardiopulmonary bypass. ABP transfusion (7-day) and in-hospital resource utilization (28-day) costs were evaluated and a multivariable net benefit regression model built for the full sample and predefined subgroups.
Results
Patient level costs for 495 patients were evaluated (mean [SD] age 59.2 [15.4] years and 69.3% male.) Consistent with FIBERS, ABP transfusions and adverse events were similar in both treatment groups. Median (IQR) total 7-day ABP cost was CAD $2280 (US dollars [USD] $1697) (CAD $930 [USD $692]-CAD $4970 [USD $3701]) in the fibrinogen concentrate group and CAD $2770 (USD $1690) (IQR, CAD $1140 [USD $849]-CAD $5000 [USD $3723]) in the cryoprecipitate group. Median (interquartile range) total 28-day cost was CAD $38 180 (USD $28 431) $(IQR, CAD $26 350 [USD $19 622]-CAD $65 080 [USD $48 463]) in the fibrinogen concentrate group and CAD $38 790 (USD $28 886) (IQR, CAD $26 180 [USD $19 495]-CAD $70 380 [USD $52 409]) in the cryoprecipitate group. After exclusion of patients who were critically ill before surgery (11%) due to substantial variability in costs, the incremental net benefit of fibrinogen concentrate vs cryoprecipitate was positive (probability of being cost-effective 86% and 97% at $0 and CAD $2000 (USD $1489) willingness-to-pay, respectively). Net benefit was highly uncertain for nonelective and patients with critical illness.
Conclusions and Relevance
Fibrinogen concentrate is cost-effective when compared with cryoprecipitate in most bleeding adult patients who underwent cardiac surgery with acquired hypofibrinogenemia requiring fibrinogen replacement. The generalizability of these findings outside the Canadian health system needs to be verified.
Introduction
Excessive bleeding related to acquired hypofibrinogenemia is a serious complication of cardiac surgery. Fibrinogen levels can be supplemented during surgery using cryoprecipitate or fibrinogen concentrate (FC).1,2 FC is purified from human plasma, is pathogen reduced, has a standardized amount of fibrinogen, and benefits from a long shelf life.3,4 Cryoprecipitate is also plasma derived but is nonpurified, contains a variable amount of fibrinogen and other coagulation factors, and has a shorter shelf life both frozen and after thawing.5 There is limited high-quality evidence comparing effectiveness, safety, and cost-effectiveness of these 2 products.6 Current guideline recommendations on the management of acquired hypofibrinogenemia are inconsistent2,7,8 and reflect differences in practice preferences, product availability, and costs in different countries.
The Fibrinogen Replenishment in Surgery (FIBERS) randomized clinical trial (RCT) examined the efficacy and safety of FC (Fibryga; Octapharma AG) vs cryoprecipitate in bleeding adult patients of cardiac surgery with acquired hypofibrinogenemia.9 In total, 827 patients from 11 Canadian centers were enrolled between February 2017 and November 2018 and randomized (1:1) to either FC or cryoprecipitate.9 This study found FC to be noninferior to cryoprecipitate with regard to the number of allogeneic blood products (ABP) transfused within 24 hours postcardiopulmonary bypass (CPB) (mean ratio = 0.96 with 1-sided 97.5% CI from −∞ to 1.09; P < .001 for noninferiority) but did not include a cost-effectiveness analysis.9
Evidence on cost-effectiveness of FC vs cryoprecipitate use is limited.6 A 2016 US-based economic evaluation used a cost-minimization model along with a survey of US Transfusion Medicine fellowship directors to evaluate FC vs cryoprecipitate in patients with acquired bleeding from the transfusion service perspective.10 The study found that the FC was more expensive than cryoprecipitate by CAD $976 (USD $727) to CAD $1303 (USD $970) per patient. To be economically competitive, a cost reduction of 44% from the current cost of FC to CAD $414 (USD $308) per gram would be required, or a saving of 0.25 to 0.66 intensive care unit (ICU) days would need to be made for each survivor. Among survey respondents, 96.7% reported not using FC because of cost (30%), off-label use (27%), and insufficient evidence (20%).10 A 2014 report by the Canadian Agency for Drugs and Technologies in Health evaluated 1 brand of FC (RiaSTAP; CSL Behring Canada) against nonfibrinogen options, finding a lack of relevant head-to-head effectiveness and cost-effectiveness studies.11 Considering the lack of real-world economic evaluation studies, the present study aimed to determine the cost-effectiveness of FC vs cryoprecipitate in adult cardiac surgery patients in need of fibrinogen replacement to treat post–CPB active bleeding related to acquired hypofibrinogenemia.
Methods
Study Design
This study was a within-trial economic evaluation of data collected as part of the FIBERS trial.9 Reporting followed Consolidated Health Economic Evaluation Reporting Standards (CHEERS) reporting guidelines. The study protocol was approved by each participating center’s research ethics board, including approval of delayed written consent in accordance with the Canadian Tri-Council Policy Statements on ethical conduct of research involving humans.9
The FIBERS RCT (NCT03037424), described in full elsewhere, enrolled patients who underwent cardiac surgery with CPB from 11 centers across 4 Canadian provinces.9,12 Treatment administration protocol is reported in eMethods 1 in the Supplement. This economic evaluation included costing data from Ontario-based hospitals only, as these were the only sites where costing data was available. Economic evaluation was conducted from the hospital perspective, capturing all medical costs incurred by patients from hospital admission. The cost of ABPs as provided by Canadian Blood Services was also taken into consideration, although these products are provided to hospitals without charge and are reimbursed by the Ontario provincial government. The time horizon, the fixed time period for economic evaluation, extended to the date of discharge or 28 days postsurgery, whichever occurred first. The date of surgery corresponded to the patient’s randomization date. Neither costs nor health outcomes were discounted due to the short time horizon. All costs were reported in 2020 Canadian dollars after applying Consumer Price Index for Health and Personal care, as appropriate.13
Effectiveness Outcomes for Cost-effectiveness Analysis
The primary effectiveness outcome for the cost-effectiveness analysis was the number of ABPs (red blood cells [RBCs], platelets, and plasma) administered within 7 days post-CPB. The secondary effectiveness outcome was ABPs administered within 24 hours of CPB. Though not used in the cost-effectiveness analysis, other clinical outcomes within 28 days of surgery, such as ICU stay, duration of mechanical ventilation, and hospitalization, as well as adverse events (AEs) including bleeding and death are also reported here for completeness. Bleeding categories were based on the validated universal definition of perioperative bleeding in cardiac surgery.14
Costs Evaluation
To evaluate costs, a bottom-up costing approach was used to conduct patient-level costing.15 All Ontario hospitals use the standardized Management Information Systems Patient Costing Methodology (known also as activity-base costing).12 Case costing data were requested from the finance departments of all Ontario hospitals that participated in the FIBERS trial. Per-patient medical costs over their hospital stay (up to 28 days postsurgery) including direct (eg, fixed, and variable service-recipient and nonservice recipient) and indirect (eg, administrative services, systems support services) costs were summarized by resource utilization category (eg, operating room [OR], ICU, nursing, pharmacy). Blood product costs included the cost of initial treatment with FC or cryoprecipitate and costs up to 7 days post-CPB for ABPs, recombinant factor VIIa (rFVIIa), and prothrombin complex concentrate. Unit costs for all products were provided by the Canadian Blood Services (eTable 1 in the Supplement). To account for unused, returned, or discarded products (ie, wastage), the costs were inflated by 1.02 for RBCs, 1.19 for platelets, 1.07 for plasma, and 1.07 for cryoprecipitate transfusions (as per data on wastage provided by Canadian Blood Services). Exploratory analyses were conducted to understand large variations in total costs by treatment site and predefined subgroups with expected variation in costs.
Cost-effectiveness Analysis
Baseline characteristics of the 2 treatment groups were compared using the standardized mean difference (SMD), ie, the number of standard deviations difference between groups16 with values more than 0.1 indicating imbalance. Usage of ABPs was compared between groups using the Wilcoxon rank sum test. For each resource utilization category, mean costs in each group and differences in means between groups were calculated with bootstrap 95% CIs based on 10 000 replicates.
The cost-effectiveness analysis was conducted using incremental net monetary benefit (INB) and net benefit regression (NBR) approaches.17 INB was used to provide an estimate of cost-effectiveness by subtracting the incremental cost from the incremental effect valued in willingness-to-pay (WTP) for a unit of effectiveness.18
The outcome of the analysis was the INB of FC compared with cryoprecipitate at a defined WTP. In this study, WTP was defined as the monetary value that a decision maker would be willing to pay for a 1 unit decrease in use of ABPs. Since this value is not established in the literature, the WTP was varied across a reasonable range, from $0 to CAD $3000 (USD $2234). FC was deemed cost-effective if the INB was greater than 0 when compared with cryoprecipitate at the defined WTP.19
To assess the effect of key variables on cost-effectiveness, NBR was used to build a multivariable model accounting for covariates including treatment group, site, baseline covariates with SMD more than 0.1, and covariates for the predefined subgroups and the interaction term between treatment group and subgroup to conduct an adjusted cost-effectiveness analysis (eMethods 2 in the Supplement). We considered 2 predefined subgroups, critical vs noncritical illness status, and elective vs nonelective surgery that were found important in the analysis of the primary effectiveness outcome in FIBERS RCT. Patients were considered (by blinded adjudication at the time of initial data collection) to be in a critical preoperative state if undergoing emergency surgery and having ventricular tachycardia/fibrillation or cardiac arrest; preoperative cardiac massage; preoperative ventilation before surgery; preoperative inotropes or ventricular assist devices; preoperative acute kidney failure (anuria or oliguria less than 10 mL/h); or acute aortic dissection.9 A cost-effectiveness acceptability curve was used to report the uncertainty of cost-effectiveness at various WTP values.
Results
Resource utilization and costing data were available for 4 of the 7 Ontario hospitals in the FIBERS trial, with a total sample of 507 treated patients who provided consent. After excluding 12 patients with missing costing data, the final sample for cost-effectiveness analyses included 495 patients representing 495 of 735 patients (67.3%) in the primary effectiveness analysis set of the FIBERS study.
Baseline Characteristics
Baseline characteristics of the study population are shown in Table 1. Balance between treatment groups was maintained for most characteristics, the notable exception being that there were more patients in a critical state before surgery in the FC group (37 of 251 [14.7%]) than in the cryoprecipitate group (17 of 244 [7.0%]), as also observed in the primary analysis of the FIBERS study. The distribution of surgical factors also differed between study sites (eTable 2 in the Supplement).
Table 1. Characteristics of the Study Population at Baseline.
| Characteristics | No. (%) | SMD | |
|---|---|---|---|
| FC (n = 251) | Cryoprecipitate (n = 244) | ||
| Age, mean (SD), y | 59.2 (15.4) | 59.4 (15.8) | 0.014 |
| Sex | |||
| Male | 174 (69.3) | 170 (69.7) | 0.008 |
| Female | 77 (30.7) | 74 (30.3) | |
| Race | |||
| White | 179 (71.3) | 164 (67.2) | 0.101 |
| Asian | 41 (16.3) | 49 (20.1) | |
| Black, American Indian, and othera | 31 (12.4) | 31 (12.7) | |
| Treatment site | |||
| A | 183 (72.9) | 172 (70.5) | 0.060 |
| B | 31 (12.4) | 31 (12.7) | |
| C | 20 (8.0) | 22 (9.0) | |
| D | 17 (6.8) | 19 (7.8) | |
| NYHA classification | |||
| I | 65 (25.9) | 70 (28.7) | 0.091 |
| II | 81 (32.3) | 81 (33.2) | |
| III | 76 (30.3) | 70 (28.7) | |
| IV | 29 (11.6) | 23 (9.4) | |
| Left ventricular function | |||
| Good (EF>50%) | 185 (76.1) | 182 (76.5) | 0.231 |
| Moderate (EF 31–50%) | 26 (10.7) | 38 (16.0) | |
| Poor (EF 21–30%) | 18 (7.4) | 9 (3.8) | |
| Very poor (EF<21%) | 14 (5.8) | 9 (3.8) | |
| Comorbid conditions | |||
| Hypertension | 149 (59.4) | 152 (62.3) | 0.060 |
| Congestive heart failure | 69 (27.5) | 61 (25.0) | 0.057 |
| Atrial fibrillation | 49 (19.5) | 44 (18.0) | 0.038 |
| Diabetes | 50 (19.9) | 45 (18.4) | 0.038 |
| Chronic lung disease | 30 (12.0) | 19 (7.8) | 0.140 |
| Stroke/TIA | 32 (12.7) | 31 (12.7) | 0.001 |
| Preoperative laboratory values, median (IQR) | |||
| Creatinine, μmol/L | 84.0 (72.0-102.0) | 86.0 (70.0-103.0) | 0.151 |
| Hemoglobin, g/dL | 135.0 (122.2-146.8) | 137.0 (123.0-149.0) | 0.058 |
| Platelet count, 103/μL | 195.0 (157.0-235.0) | 186.0 (155.0-224.0) | 0.026 |
| INR | 1.0 (1.0-1.1) | 1.0 (1.0-1.1) | 0.157 |
| Surgical factor | |||
| Critically ill before surgeryb | 37 (14.7) | 17 (7.0) | 0.252 |
| Nonelective surgery | 83 (33.1) | 73 (29.9) | 0.068 |
| Complex surgeryc | 176 (70.1) | 178 (73.0) | 0.063 |
| Type of surgical procedure | |||
| Aortic valve procedure (AV repair/replace) | 78 (31.1) | 89 (36.5) | 0.114 |
| Surgery on aorta (ascending/arch/descending) | 111 (44.2) | 97 (39.8) | 0.091 |
| CABG surgery | 90 (35.9) | 93 (38.1) | 0.047 |
| MV procedure (MV repair/replace) | 40 (15.9) | 43 (17.6) | 0.045 |
| CPB duration, median (IQR), min | 140 (100-200) | 134 (102-199) | 0.049 |
Abbreviations: AV, atrial valve; CABG, coronary artery bypass graft; CPB, cardiopulmonary bypass; EF, ejection fraction; FC, fibrinogen concentrate; INR, international normalized ratio; IQR, interquartile range; MV, mitral valve; NYHA, New York Heart Association; SMD, standardized mean difference; TIA, transient ischemic attack.
SI conversion factor: To convert μmol/L to mg/dL, multiply by 0.95; to convert μL to L, multiply by 1.
Other races marked as unknown or not applicable.
Critical illness was determined by blinded adjudication on patients who underwent emergency surgery and had any of the following conditions: (1) ventricular tachycardia or fibrillation or cardiac arrest, (2) preoperative cardiac massage, (3) preoperative ventilation before anesthetic room, (4) hemodynamic support requiring preoperative inotropes or ventricular assist devices, (5) preoperative acute kidney failure (anuria or oliguria less than 10 mL/h), (6) acute aortic dissection.
Procedures other than CABG surgery only, single valve only, or repair of atrial septal defect only.
Effectiveness Outcomes
Table 2 provides a summary of ABP usage within the first 24 hours and 7 days post-CBP, with approximately 90% used within the first 24 hours. Consistent with the primary analysis of the FIBERS trial, there were numerically fewer ABP transfusions in the FC group vs the cryoprecipitate group, a pattern that was apparent across blood product categories, within the first 24 hours and 7 days postsurgery. Overall, there was substantial variation in patients’ individual ABP usage for each treatment group (eFigure 1 in the Supplement). No differences were observed between the groups regarding the use of prothrombin complex concentrate and rFVIIa (eTable 3 in the Supplement). Incidence of AEs, severe or massive bleeds, and deaths were similar across the 2 treatment groups at 28-day follow-up (eTable 4 in the Supplement), with no significant differences observed in ICU stay, duration of mechanical ventilation, or hospitalization.
Table 2. Primary and Secondary Effectiveness Outcomes: Usage of Blood Components.
| Outcomes | FC (n = 251) | Cryoprecipitate (n = 244) | P valueb | ||
|---|---|---|---|---|---|
| Mean (SD) | Median (IQR) | Mean (SD) | Median (IQR) | ||
| Cumulative ABPa transfused within 24 h after CPB | 14.2 (16.8) | 9 (4-20) | 15.5 (14.8) | 12 (5-21) | .03 |
| RBC transfusions within 24 h after CPB | 2.9 (4.2) | 2 (0-4) | 3.1 (3.7) | 2 (0-5) | .13 |
| Platelet transfusions within 24 h after CPB | 8.0 (8.2) | 8 (4-12) | 8.8 (7.2) | 8 (4-12) | .02 |
| Plasma transfusions within 24 h after CPB | 3.3 (5.8) | 2 (0-4) | 3.6 (5.2) | 2 (0-4) | .18 |
| Cumulative ABP transfused within 7 d after CPB (primary) | 16.2 (19.2) | 10 (4-21) | 17.1 (16.6) | 13 (6-22) | .06 |
| RBC transfusions within 7 d after CPB | 4.2 (6.0) | 2 (1-6) | 4.2 (5.1) | 3 (1-6) | .36 |
| Platelet transfusions within 7 d after CPB | 8.3 (8.9) | 8 (4-12) | 9.1 (7.6) | 8 (4-12) | .02 |
| Plasma transfusions within 7 d after CPB | 3.7 (6.2) | 2 (0-4) | 3.8 (5.6) | 2 (0-4) | .24 |
Abbreviations: ABP, allogeneic blood products; CPB, cardiopulmonary bypass; FC, fibrinogen concentrate; IQR, interquartile range; RBC, red blood cell.
Units of ABPs counted as follows: each RBC unit = 1 unit, each 250 mL plasma unit = 1 unit, each 500 mL plasma unit = 2 units, each platelet dose = 4 units.
Usage of ABPs was compared between groups using the Wilcoxon rank sum test.
Resource Utilization and Costs
Median (interquartile range [IQR]) total 7-day ABP cost was CAD $2280 (USD $1698) (IQR, CAD $930 [USD $693]-CAD $4970 [USD $3701]) CAD in the fibrinogen concentrate group and $2770 (USD $2063) (IQR, CAD $1140 [USD $849]-CAD $5000 [USD $3723]) in the cryoprecipitate group. Similar total costs were seen across the 2 treatment groups at 28 days postsurgery (ie, sum of treatment, 7-day ABP, and 28-day in-hospital resource costs) (Table 3), but there was wide between-patient variation. In-hospital resource utilization costs accounted for around 93% of the total per-patient 28-day costs, with the rest being attributable to ABPs. Approximately one-third of the total in-hospital resource utilization cost was attributable to ICU stay (36%), followed by OR use (26%) and nursing (9%).
Table 3. 7-Day ABP Transfusions and 28-Day In-Hospital Health Care Resource Utilization Costs.
| Resource utilization | All patients | Patients with noncritical illness | ||||
|---|---|---|---|---|---|---|
| FC (n = 251) | Cryoprecipitate (n = 244) | Incremental cost, mean (95% CI, CAD $)a,b | FC (n = 214) | Cryoprecipitate (n = 227) | Incremental cost, mean (95% CI, CAD $) | |
| Treatment, mean (SD) | 1970 (780) | 1600 (850) | 370 (220 to 510) | 1860 (600) | 1560 (820) | 300 (160-430) |
| Median (IQR) | 1710 (1690 to 1710) | 1270 (1260-1270) | 1710 (1690-1710) | 1270 (1260-1270) | ||
| 7-d ABP transfusions | ||||||
| Cumulative ABP transfusions, mean (SD) | 4170 (5340) | 4280 (5280) | −110 (−1040 to 840) | 3280 (4330) | 3960 (4910) | −670 (−1520 to 170) |
| Median (IQR) | 2280 (930 to 4970) | 2770 (1140-5000) | 1950 (890-3770) | 2600 (950-4600) | ||
| RBC transfusions, mean (SD) | 1890 (2660) | 1860 (2280) | 30 (−400 to 470) | 1510 (2220) | 1790 (2230) | −280 (−690 to 140) |
| Median (IQR) | 900 (440 to 2450) | 1330 (450-2660) | 900 (440-1790) | 1330 (450-2240) | ||
| Platelet transfusions, mean (SD) | 650 (700) | 710 (600) | −60 (−170 to 50) | 510 (550) | 680 (560) | −160 (−270 to −60) |
| Median (IQR) | 620 (310 to 940) | 620 (310-950) | 320 (0-640) | 620 (310 to 940) | ||
| Plasma transfusions, mean (SD) | 430 (710) | 450 (650) | −20 (−140 to 100) | 350 (670) | 430 (650) | −80 (−200 to 50) |
| Median (IQR) | 230 (0-480) | 240 (0-500) | 230 (0-460) | 240 (0-480) | ||
| rFVIIa transfusions, mean (SD) | 510 (1590) | 570 (2020) | −60 (−380 to 260) | 370 (1380) | 430 (1740) | −60 (−350 to 210) |
| Median (IQR) | 0 (0 to 0) | 0 | 0 | 0 | ||
| PCC transfusions, mean (SD) | 370 (650) | 330 (610) | 40 (−80 to 150) | 280 (560) | 300 (570) | −20 (−130 to 80) |
| Median (IQR) | 0 (0590) | 0 (0-590) | 0 | 0 (0-590) | ||
| 28-d In-hospital resource use | ||||||
| Total in-hospital costs, mean (SD) | 54 640 (61 160) | 53 320 (61 770) | 1320 (−9740 to 12160) | 46 170 (50 640) | 48 880 (46 470) | −2710 (−11 660 to 6640) |
| Median (IQR) | 33 510 (23 370 to 56 910) | 33 600 (22 940-60 120) | 31 310 (22 420-51 470) | 32 090 (22 320-54 540) | ||
| ICU, mean (SD) | 20 510 (32 000) | 18 120 (23 590) | 2380 (−2310 to 7460) | 16 410 (25 710) | 17 340 (23 600) | −930 (−5350 to 3890) |
| Median (IQR) | 9340 (3700 to 20 670) | 8700 (4220-19 960) | 8070 (3700-18 430) | 8310 (3880-18 470) | ||
| OR, mean (SD) | 13 050 (21 760) | 14 550 (37 840) | −1500 (−7470 to 3400) | 11 020 (14 800) | 12 570 (19 600) | −1550 (−5020 to 1590) |
| Median (IQR) | 9280 (6180 to 12 750) | 9310 (6550-13 170) | 8850 (6170-12 180) | 9290 (6570-13 050) | ||
| Nursing, mean (SD) | 4630 (3670) | 5480 (4900) | −850 (−1590 to −110) | 4760 (3280) | 5140 (3990)) | −380 (−1050 to 280) |
| Median (IQR) | 3900 (2680 to 5570) | 4260 (3010-5840) | 4040 (2930-5640) | 4250 (3020-5640 | ||
| RT, mean (SD) | 3750 (5930) | 3480 (4400) | 270 (−650 to 1220) | 3060 (4530) | 3380 (4490) | −330 (−1160 to 540) |
| Median (IQR) | 1970 (1380 to 3140) | 2000 (1470-3520) | 1850 (1340-2720) | 1940 (1460-3000) | ||
| Pharmacy, mean (SD) | 2730 (4850) | 2270 (3270) | 450 (−240 to 1210) | 2330 (4640) | 2000 (2730) | 330 (−340 to 1070) |
| Median (IQR) | 1030 (700 to 2500) | 1040 (710-2440) | 900 (650-1900) | 1010 (690-2010) | ||
| Diagnostics/testing, mean (SD) | 3200 (3980) | 3140 (3120) | 70 (−560 to 720) | 2660 (3440) | 2940 (2900) | −280 (−870 to 320) |
| Median (IQR) | 1900 (1110 to 3600) | 2090 (1260-3950) | 1660 (990-2810) | 1940 (1260-3430) | ||
| Allied health, mean (SD) | 830 (1100) | 960 (1310) | −130 (−340 to 80) | 710 (970) | 850 (1140) | −140 (−340 to 60) |
| Median (IQR) | 360 (170 to 940) | 390 (200-1110) | 320 (160-770) | 350 (180-1040) | ||
| Other, mean (SD) | 5940 (11 920) | 3120 (2120-5570) | 620 (−1080 to 2630) | 5220 (11 930) | 4650 (6380) | 570 (−1020 to 2510) |
| Median (IQR) | 3070 (2120 to 6140) | 3120 (2120-5570) | 2840 (2040-5380) | 3000 (2070-5210) | ||
| Total costs,c mean (SD) | 60 780 (64 910) | 59 200 (64 510) | 1580 (−10 000-13 040) | 51 320 (53 370) | 54 400 (49 790) | −3080 (−12 640 to 6710) |
| Median (IQR) | 38 180 (26 350 to 65 080) | 38 790 (26 180-70 380) | 35 390 (25 920-56 850) | 37 890 (25 700-62 030) | ||
Abbreviations: ABP, allogeneic blood product; FC, fibrinogen concentrate; ICU, intensive care unit; IQR, interquartile range; OR, operating room; PCC, prothrombin complex concentrate; RBC, red blood cell; rFVIIa, recombinant activated factor VII; RT, respiratory therapy.
95% CIs were calculated based on percentiles of 10 000 bootstrap replicates.
To convert to US dollars, multiply by 1.26.
Sum of treatment, 7-day transfusions, and 28-day in-hospital health care costs.
To better understand the sources of large variations in health resource utilization, subgroup analyses were performed. As per eTable 5 in the Supplement, 95% CIs of mean cost differences in all subgroup comparisons between treatment groups included 0. However large differences were observed in health care costs stratified by treatment site (eFigure 2 in the Supplement) and by surgical factors (eTable 5 in the Supplement). Patients who died had overall higher total and ABP costs compared with other patients. Patients with baseline critical illness (n = 54) had almost double mean health care and ABP costs and wide between-patient variability with the CI for the mean incremental cost of fibrinogen concentrate vs cryoprecipitate consistent with both an enormous per-patient savings of CAD $93 980 (USD $69 984) and excess cost of CAD $57 160 (USD $42 565) (eTable 5 in the Supplement). Further evaluation of in-hospital resource utilization between treatment groups by critical illness status found no significant differences in mean costs of individual components, except for nursing, which had higher costs in critically ill patients for the cryoprecipitate group (Table 3 and eTable 6 in the Supplement).
Cost-effectiveness Analyses (Unadjusted)
Bootstrapped pairs of mean incremental cost and mean incremental effectiveness (ie, units of ABP use) for all patients are shown in eFigure 3 in the Supplement. Values plotted on the cost-effectiveness plane show a wide distribution, with a mean incremental cost of CAD $1580 (USD $1177) and a mean incremental effectiveness of 0.9 units for FC vs cryoprecipitate.
The bootstrapped cost-effectiveness pairs by critical illness status are shown in Figure 1. The larger sample of patients with noncritical illness (n = 441) shows less variability for the mean incremental costs and benefits, with a mean incremental cost of −CAD $3030 (−USD $2256) and a mean incremental effectiveness of 3.3 units for FC vs cryoprecipitate. Moreover, 73% of paired values were distributed in the lower right quadrant indicating that FC was more effective and less costly (ie, dominant). The proportion of paired bootstrapped values in patients with noncritical illness below the WTP of CAD $2000 (USD $1489) was 91%. There was substantial variability in the small sample of patients with critical illness (n = 54). Similar findings were found when comparing patients by elective status, with 70% of paired values being in the dominant quadrant among elective patients and observing substantial variability among nonelective patients (n = 102).
Figure 1. Cost-effectiveness Plane, Fibrinogen Concentrate (FC) vs Cryoprecipitate, by Critical Illness Status.
Cost-effectiveness plane of bootstrap cost-effectiveness values based on 10 000 replicates. The blue dot represents the mean incremental cost (−CAD $3030 [USD $2259] and mean incremental effectiveness [3.3 units]) when comparing FC against cryoprecipitate for noncritically ill patients. The orange dot represents the mean incremental cost (CAD −$8390 [USD $6254]) and mean incremental effectiveness (−7.4 allogeneic blood products) when comparing fibrinogen concentrate against cryoprecipitate for critically ill patients. The dashed line represents the hypothetical willingness-to-pay threshold of CAD $2000 (USD $1489) to avoid 1 unit of allogeneic blood products transfusion.
Net Benefit Regression
Multivariable NBR models were built accounting for covariates to allow INB to differ by severity. In noncritically ill patients, the ABP transfusions and the overall mean costs were lower in the FC group (Table 3). At a WTP of $0, the net benefit was positive (INB = CAD $4820 [USD $3589]; 95% CI, –CAD $4270 [−USD $3180]-13 410 [USD $9986]) and increased to CAD $12 050 (USD $8973) (95% CI, CAD $–440 to CAD 24 600 [USD $18319]) for a WTP of CAD $2000 [USD $1489] (Figure 2A). At WTP of 0, the probability of FC being cost-effective (ie, cost-saving) was 86%, while at a WTP of CAD $2000 the probability of FC being cost-effective was 97% (Figure 2, B; eTable 7 in the Supplement). The critically ill group made up a very small subset of patients with substantially higher and variable costs. Therefore, estimation of net benefit was imprecise for these patients.
Figure 2. Incremental Net Benefit and Cost-effectiveness Acceptability Curve by Critical Illness Status.
In adjusted NBR, fibrinogen was cost-effective among elective patients (probability of cost-effectiveness 86% and 97% for WTP of 0 and CAD $2000 (USD $1489), respectively) and highly uncertain and not cost-effective (47% and 41%) in nonelective patients (eFigure 4, eTable 7 in the Supplement). After excluding critically ill patients, the probability of cost-effectiveness was 97% for elective and 78% for nonelective patients.
Discussion
This study was a real-world within-trial cost-effectiveness evaluation of FC vs cryoprecipitate that included 67.3% of patients in the primary effectiveness analysis of the FIBERS clinical trial.9 High-quality evidence comparing effectiveness, safety, and cost-effectiveness of FC vs cryoprecipitate is currently limited.6,10 This study provides a new perspective on the results on the FIBERS trial showing FC to be cost-effective in comparison with cryoprecipitate for most bleeding adult cardiac surgery patients requiring fibrinogen supplementation.
As in the primary analysis of the FIBERS study, this analysis showed nonsignificantly lower numbers of ABP were required by the FC group 24 hours and 7 days postsurgery compared with the cryoprecipitate group, with primary and secondary effectiveness outcomes consistent with those from the full trial.9 Overall, the mean ABP costs were similar across both treatment groups and the difference in the costs between the 2 groups was largely due to hospital costs. Average 28-day in-hospital costs were high due to the complexity of surgical procedures (aortic valve replacement, surgery on aorta, coronary artery bypass, etc) and patient comorbidities. As expected, ICU, OR, and nursing costs were the main cost drivers of in-hospital resource utilization costs. The total average total costs varied by site, emergency surgery status, surgical complexity, critical illness, and vital status at discharge. Moreover, heterogeneity in costs translated into a large uncertainty in cost-effectiveness estimates for the overall sample limiting any meaningful conclusion.
The NBR approach allowed modeling of the net benefit as a function of treatment group and covariates (site, preoperative critical illness status, and the interaction of treatment and critical illness) and helped to explain some of the variation in net benefit. Predictably, patients with critical illness status required extensive perioperative support and care, and one-third of these patients died during the first admission. Patients with critical illness and those who died required almost double the health care resource utilization and costs of other patients. Among noncritically ill patients (89% of patients), FC was shown to be more effective and less costly than cryoprecipitate (with 73% of bootstrapped samples in the dominant quadrant and 86% probability of being cost-effective at TTP of $0) while results in the critically ill subgroup remained highly variable. FC was cost-effective in elective patients while it went from being not cost-effective to cost-effective after excluding patients with critical illness from the nonelective surgery group.
A more rigorous approach in economic evaluation studies is based on using quality-adjusted life-years (QALYs) as an effectiveness estimate and calculating INB for common WTP thresholds, such as CAD $50 000 (USD $37 233) or CAD $100 000 (USD $74 467) per QALY.20 Considering the short duration of the intervention and low AE rates potentially related to fibrinogen concentrate, cryoprecipitate, or other ABP that could have chronic, lifetime implications, QALYs were not evaluated in this study. For this analysis, a WTP range from $0 to CAD $3000 (USD $2234) (per ABP unit avoided was used, with CAD $2000 (USD $1489) as a hypothetical threshold in cost-effectiveness planes. However, the true WTP that a patient or a decision maker may be willing to pay to avoid 1 unit of ABP transfusion is unknown. A systematic review of studies reporting cost of RBC transfusion in European countries reported that “the population-weighted mean cost of transfusing 2 units of blood was €877.69 (USD $923).”21 A US-based study that used an activity-based model to estimate the cost of RBC transfusion from a hospital perspective and considered the costs of collecting, processing, and transfusing blood, as well as monitoring and managing of acute transfusion-related reactions, estimated the mean cost per RBC unit equal to $761 (SD, $294).22 The cost of ABP transfusion, however, may differ from the WTP to avoid additional ABP transfusion. A 1998 study by Lee et al23 estimated the WTP for allogeneic blood donation (ABD) among 432 patients scheduled for ABD for various planned surgical procedures. Among patients (n = 202) who received information about the risks of allogeneic blood transfusions (eg, risk of hepatitis, HIV, transfusion reaction) the median WTP for ABD (to avoid additional ABP transfusion) was $1097 compared with $1908 when no such information was provided.23 Taking into consideration the decreased risk of ABP transfusion complications over the last 2 decades, more studies are needed in this area.
Limitations
This economic evaluation captured the hospital perspective while also considering costs of ABPs paid by the Ontario provincial government. Since the ABP costs were only 7% of the total costs, this analysis largely reflects the hospital perspective. In Canada, under the universal health care system, a fixed global (annual) budget is the most common model for individual hospital funding. About 72% of study sample came from a single, quaternary academic hospital. Therefore, the generalizability of the current cost-effectiveness analysis could be limited for other Canadian provinces, for resource-constrained settings, and for countries with different health care models and costs. Patient follow-up in this study was limited to 28 days after surgery; thus, long-term differences in transfusion-related pathogen safety/AEs as reported by hemovigilance systems24 were not captured and therefore were not used for cost assessment. A health care perspective was not used since physician billings related to surgical procedures, management of AEs, potential readmissions, diagnostic services, and other resource utilization costs covered by the provincial health care insurance plan during the study time horizon were not captured. A societal perspective that considered hospital, health care system, and patient costs was also not appropriate since the health care and patient perspectives were not captured by this study. Considering the randomization (potentially balancing physician billing and patient costs) and the potential short effect of the intervention on health care utilization and outcomes, a hospital perspective with a 28-day time horizon was deemed a reasonable approach.
Conclusions
In conclusion, this study showed FC to be cost-effective when compared with cryoprecipitate in most bleeding adult cardiac surgery patients with acquired hypofibrinogenemia requiring fibrinogen replacement. While the FIBERS trial provided strong evidence on the noninferiority of FC vs cryoprecipitate, this economic evaluation provides an additional perspective on its cost-effectiveness. The generalizability of these findings outside the Canadian health system needs to be verified.
eMethods 1. Hypofibrinogenemia treatment protocol
eMethods 2. Net Benefit Regression
eTable 1. Unit Costs of Blood Products and Plasma Protein Products, CAD
eTable 2. Surgical Factors and Effectiveness Outcomes by Study Site
eTable 3. Other Blood Products Within 7 d of CPB
eTable 4. Other Clinical Outcomes at 28 days Follow-up
eTable 5. Total 28 day Healthcare and 7 day Blood Product Costs Per Patient by Select Subgroups, CAD
eTable 6. Critically-ill Patients: 28 day In-hospital healthcare resource utilization costs, CAD
eTable 7. INB of Using FC vs. Cryoprecipitate
eFigure 1. ABP Usage Within 7 day of Termination of CPB
eFigure 2. Total Healthcare Costs by Treatment Site and Treatment Group
eFigure 3. Cost-effectiveness Plane, FC vs. Cryoprecipitate, all patients
eFigure 4. Incremental Net Benefit (A, C) and Cost-effectiveness Acceptability Curve (B, D), by Elective Status
References
- 1.Spahn DR, Bouillon B, Cerny V, et al. Management of bleeding and coagulopathy following major trauma: an updated European guideline. Crit Care. 2013;17(2):R76. doi: 10.1186/cc12685 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tibi P, McClure RS, Huang J, et al. STS/SCA/AmSECT/SABM Update to the clinical practice guidelines on patient blood management. J Extra Corpor Technol. 2021;53(2):97-124. doi: 10.1182/ject-2100053 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Martini WZ, Guzman R, Dubick MA. Stability of fibrinogen concentrate in human blood samples: an in vitro study. Mil Med. 2018;183(suppl 1):183-188. doi: 10.1093/milmed/usx176 [DOI] [PubMed] [Google Scholar]
- 4.Novak A, Stanworth SJ, Curry N. Do we still need cryoprecipitate? cryoprecipitate and fibrinogen concentrate as treatments for major hemorrhage—how do they compare? Expert Rev Hematol. 2018;11(5):351-360. doi: 10.1080/17474086.2018.1458610 [DOI] [PubMed] [Google Scholar]
- 5.Callum JL, Karkouti K, Lin Y. Cryoprecipitate: the current state of knowledge. Transfus Med Rev. 2009;23(3):177-188. doi: 10.1016/j.tmrv.2009.03.001 [DOI] [PubMed] [Google Scholar]
- 6.Jensen NH, Stensballe J, Afshari A. Comparing efficacy and safety of fibrinogen concentrate to cryoprecipitate in bleeding patients: a systematic review. Acta Anaesthesiol Scand. 2016;60(8):1033-1042. doi: 10.1111/aas.12734 [DOI] [PubMed] [Google Scholar]
- 7.Boer C, Meesters MI, Milojevic M, et al. ; Task Force on Patient Blood Management for Adult Cardiac Surgery of the European Association for Cardio-Thoracic Surgery (EACTS) and the European Association of Cardiothoracic Anaesthesiology (EACTA) . 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth. 2018;32(1):88-120. doi: 10.1053/j.jvca.2017.06.026 [DOI] [PubMed] [Google Scholar]
- 8.Spahn DR, Bouillon B, Cerny V, et al. The European guideline on management of major bleeding and coagulopathy following trauma: fifth edition. Crit Care. 2019;23(1):98. doi: 10.1186/s13054-019-2347-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Callum J, Farkouh ME, Scales DC, et al. ; FIBRES Research Group . Effect of fibrinogen concentrate vs cryoprecipitate on blood component transfusion after cardiac surgery: The FIBRES randomized clinical trial. JAMA. 2019;322(20):1966-1976. doi: 10.1001/jama.2019.17312 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Okerberg CK, Williams LA III, Kilgore ML, et al. Cryoprecipitate AHF vs. fibrinogen concentrates for fibrinogen replacement in acquired bleeding patients—an economic evaluation. Vox Sang. 2016;111(3):292-298. doi: 10.1111/vox.12417 [DOI] [PubMed] [Google Scholar]
- 11.Blackhouse G, Assasi N, Campbell K, Goeree R. Human fibrinogen concentrate (RiaStap): comparative cost-effectiveness evaluation. Accessed November 22, 2022. https://www.cadth.ca/sites/default/files/pdf/OB0001_%20RiaSTAP_e.pdf
- 12.Canadian Institute for Health Information (CIHI) . Canadian patient cost database technical document: mis patient costing methodology January 2019. Accessed November 22, 2022. https://www.cihi.ca/sites/default/files/document/mis_patient_cost_meth_en_0.pdf
- 13.Statistics Canada . Consumer price index, annual average, not seasonally adjusted. Accessed November 22, 2022. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810000501
- 14.Dyke C, Aronson S, Dietrich W, et al. Universal definition of perioperative bleeding in adult cardiac surgery. J Thorac Cardiovasc Surg. 2014;147(5):1458-1463.e1. doi: 10.1016/j.jtcvs.2013.10.070 [DOI] [PubMed] [Google Scholar]
- 15.Chapko MK, Liu CF, Perkins M, Li YF, Fortney JC, Maciejewski ML. Equivalence of two healthcare costing methods: bottom-up and top-down. Health Econ. 2009;18(10):1188-1201. doi: 10.1002/hec.1422 [DOI] [PubMed] [Google Scholar]
- 16.Yang DS. Dalton JE. A unified approach to measuring the effect size between two groups using SAS. Accessed November 22, 2022. https://support.sas.com/resources/papers/proceedings12/335-2012.pdf
- 17.Hoch JS, Rockx MA, Krahn AD. Using the net benefit regression framework to construct cost-effectiveness acceptability curves: an example using data from a trial of external loop recorders versus Holter monitoring for ambulatory monitoring of “community acquired” syncope. BMC Health Serv Res. 2006;6:68. doi: 10.1186/1472-6963-6-68 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. 4th ed. Oxford University Press; 2015. [Google Scholar]
- 19.Hoch JS, Briggs AH, Willan AR. Something old, something new, something borrowed, something blue: a framework for the marriage of health econometrics and cost-effectiveness analysis. Health Econ. 2002;11(5):415-430. doi: 10.1002/hec.678 [DOI] [PubMed] [Google Scholar]
- 20.Neumann PJ, Cohen JT, Weinstein MC. Updating cost-effectiveness—the curious resilience of the $50,000-per-QALY threshold. N Engl J Med. 2014;371(9):796-797. doi: 10.1056/NEJMp1405158 [DOI] [PubMed] [Google Scholar]
- 21.Abraham I, Sun D. The cost of blood transfusion in Western Europe as estimated from six studies. Transfusion. 2012;52(9):1983-1988. doi: 10.1111/j.1537-2995.2011.03532.x [DOI] [PubMed] [Google Scholar]
- 22.Shander A, Hofmann A, Ozawa S, Theusinger OM, Gombotz H, Spahn DR. Activity-based costs of blood transfusions in surgical patients at four hospitals. Transfusion. 2010;50(4):753-765. doi: 10.1111/j.1537-2995.2009.02518.x [DOI] [PubMed] [Google Scholar]
- 23.Lee SJ, Liljas B, Neumann PJ, Weinstein MC, Johannesson M. The impact of risk information on patients’ willingness to pay for autologous blood donation. Med Care. 1998;36(8):1162-1173. doi: 10.1097/00005650-199808000-00005 [DOI] [PubMed] [Google Scholar]
- 24.Mowla SJ, Kracalik IT, Sapiano MRP, O’Hearn L, Andrzejewski C Jr, Basavaraju SV. A comparison of transfusion-related adverse reactions among apheresis platelets, whole blood-derived platelets, and platelets subjected to pathogen reduction technology as reported to the National Healthcare Safety Network Hemovigilance Module. Transfus Med Rev. 2021;35(2):78-84. doi: 10.1016/j.tmrv.2021.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods 1. Hypofibrinogenemia treatment protocol
eMethods 2. Net Benefit Regression
eTable 1. Unit Costs of Blood Products and Plasma Protein Products, CAD
eTable 2. Surgical Factors and Effectiveness Outcomes by Study Site
eTable 3. Other Blood Products Within 7 d of CPB
eTable 4. Other Clinical Outcomes at 28 days Follow-up
eTable 5. Total 28 day Healthcare and 7 day Blood Product Costs Per Patient by Select Subgroups, CAD
eTable 6. Critically-ill Patients: 28 day In-hospital healthcare resource utilization costs, CAD
eTable 7. INB of Using FC vs. Cryoprecipitate
eFigure 1. ABP Usage Within 7 day of Termination of CPB
eFigure 2. Total Healthcare Costs by Treatment Site and Treatment Group
eFigure 3. Cost-effectiveness Plane, FC vs. Cryoprecipitate, all patients
eFigure 4. Incremental Net Benefit (A, C) and Cost-effectiveness Acceptability Curve (B, D), by Elective Status