Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 https://doi.org/10.1186/s12891-022-05775-y RESEARCH Open Access Quality of life and cost‑effectiveness analysis of topical tranexamic acid and fibrin glue in femur fracture surgery A Merchán‑Galvis1,2 , M Posso3,4 , E Canovas5, M Jordán6, X Aguilera6 and MJ Martinez‑Zapata1,7*    Abstract Background: We assessed quality of life (QoL) of patients undergoing surgery for proximal femur fracture and per‑ formed a cost-effectiveness analysis of haemostatic drugs for reducing postoperative bleeding. Methods: We analysed data from an open, multicentre, parallel, randomized controlled clinical trial (RCT) that assessed the efficacy and safety of tranexamic acid (TXA group) and fibrin glue (FG group) administered topically prior to surgical closure, compared with usual haemostasis methods (control group). For this study we conducted a cost-effectiveness analysis of these interventions from the Spanish Health System perspective, using a time horizon of 12 months. The cost was reported in $US purchasing power parity (USPPP). We calculated the incremental cost-effectiveness ratio (ICER) per QALY (quality-adjusted life-year). Results: We included 134 consecutive patients from February 2013 to March 2015: 42 patients in the TXA group, 46 in the FG group, and 46 in the control group. Before the fracture, EuroQol visual analogue scale (EQ-VAS) health ques‑ tionnaire score was 68.6. During the 12 months post-surgery, the intragroup EQ-VAS improved, but without reaching pre-fracture values. There were no differences between groups for EQ-VAS and EuroQol 5 dimensions 5 levels (EQ-5D-5L) health question‑ naire score, nor in hospital stay costs or medical complication costs. Nevertheless, the cost of one FG treatment was significantly higher (399.1 $USPPP) than the cost of TXA (12.9 $USPPP) or usual haemostasis (0 $USPPP). When comparing the cost-effectiveness of the interventions, FG was ruled out by simple dominance since it was more costly (13,314.7 $USPPP) than TXA (13,295.2 $USPPP) and less effective (utilities of 0.0532 vs. 0.0734, respec‑ tively). TXA compared to usual haemostasis had an ICER of 15,289.6 $USPPP per QALY). Conclusions: There were no significant differences between the intervention groups in terms of postoperative changes in QoL. However, topical TXA was more cost-effective than FG or usual haemostasis. Trial registration: ClinicalTrials.gov: NCT02150720. Date of registration 30/05/2014. Retrospectively registered. Keywords: Quality of life, Hip fracture, Clinical trial, Cost-utility, Fibrin glue, Tranexamic acid Background Proximal femur fracture is one of the main causes of morbidity, disability, and mortality in the elderly popula- *Correspondence: mmartinezz@santpau.cat; mj.martinez.zapata@gmail.com tion. In particular, patients with proximal femur fractures 1 Iberoamerican Cochrane Centre‑Public Health and Clinical Epidemiology, have an increased risk of mortality, which is also related IIBSant Pau, Barcelona, Spain to age, sex, dementia, and frailty, and can reach 18–33% Full list of author information is available at the end of the article per year [1]. © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 2 of 11 The worldwide incidence of hip fractures is 14.2 mil- Design and study population lion people and its prevalence is 23.6 million [2]. The age- An open, multicentre, parallel RCT was performed, standardized rate per 100,000 habitants for incidence is including patients older than 18 years, with unilateral 182.5 and for prevalence is 298.1, compared with 1990 subcapital femoral fracture, treated with a total or partial these numbers have increased by 0.4% and 7.5%, respec- hip prosthesis, who agreed to participate and gave signed tively [2]. Therefore, annually, one in every 1,000 people informed consent [5]. The protocol was approved by the in developed countries will experience a hip fracture of clinical research ethics committees of the participating the proximal femur and this incidence increases expo- hospitals, as well as by the Spanish Agency for Medicines nentially in people older than 80 years [2]. In Spain, and Health Products. The Spanish ethics committees there is an incidence of 104 cases per 100,000 inhabitants (EC) that approved the clinical trial were EC of Hospi- which means between 45,000 and 50,000 hip fractures tal Universitari Germans Trias i Pujol de Badalona, EC per year; more than 90% will be treated surgically [3]. of Hospital Universitari Mútua Terrassa, EC of Consorci One of the main complications of surgical treatment is Sanitari de Terrassa, EC of Fundació de Gestió Sanitaria bleeding, which can cause anaemia. Hence, pharmaco- Hospital de la Santa Creu i Sant Pau and, EC of Hospital logical and non-pharmacological therapeutic strategies Clinic Barcelona, Spain. The study was conducted follow- are indicated to reduce bleeding, postoperative anaemia, ing the guidelines of the Declaration of Helsinki. and transfusions. Whereas tranexamic acid (TXA) and The recruitment period was from February 2013 to fibrin glue (FG) (a haemostatic product derived from March 2015 in six Spanish hospitals. Patients were fibrinogen) are drugs showing promising results, haemo- excluded if they had a known allergy to TXA or FG, had stasis using electrocautery is the most widely-used non- multiple fractures or pathological fractures, a history pharmacological intervention [4–6]. of thromboembolic disease, ischaemic heart disease, The resources needed to treat patients with proximal thrombogenic arrhythmias or peripheral vascular dis- femur fracture are considerable and with a wide varia- ease, had a cardiovascular device, prothrombotic clotting tion between high-income countries [7]. In a systematic abnormality, were on treatment with contraceptives or review of the costs of hip fractures globally, the pooled oestrogens, or those for whom informed consent could estimate of the cost for the index hospitalization was not be obtained prior to surgery [5]. $10,075 per patient [8]. After obtaining informed consent, 161 patients who Moreover, this amount can increase up to $43,669 per met the inclusion criteria were randomly assigned to patientat the end of the first year of follow-up if health receive one of the study interventions with an allocation and social care costs are taken into account [8]. There- ratio of 1: 1: 1. The protocol was approved by the clini- fore, it is important to identify and apply interventions cal research ethics committees of the participating hos- that can reduce postoperative complications and hospi- pitals, as well as by the Spanish Agency for Medicines tal costs. and Health Products. The protocol number is registered We previously conducted a randomized controlled in the database www.​clini​caltr​ials.​gov (NCT02150720). clinical trial (RCT) that compared the safety and efficacy The study was conducted following the guidelines of the in reducing bleeding using topical TXA or FG versus the Declaration of Helsinki, and all patients or their legal rep- usual haemostasis technique in patients undergoing pros- resentatives gave signed informed consent to participate thetic hip replacement for proximal femur fracture [5, 9]. in the study before surgery. In the per protocol analysis, topical TXA showed a sig- nificant reduction in postoperative bleeding and transfu- sions compared with the alternatives. We hypothesized Interventions that topical TXA may also produce an improvement in Control group: received haemostasis with conventional QoL compared with the other therapeutic options and diathermy applied to the surgical wound surface before that it could be a saving option to implement in clinical TXA group: received TXA (Amchafibrin®, Rottafarm its closure. practice. Therefore, we assessed the QoL of the included patients undergoing surgery for proximal femur frac- SL, Valencia, Spain), 1 g in 10 mL (concentration 100 mg/ ture and performed a cost-effectiveness analysis of these mL) of solution (2 vials of 500 mg/5 mL). Topical admin- treatments. istration was performed through a special syringe that sprayed the product on the operative field before wound FG group: received 10 mL of FG (Evicel®, Omrix Biop- closure. Methods We followed the CHEERS recommendations for report- harmaceuticals NV, Diegem, Belgium), which contains ing health-economics studies [10]. mainly fibrinogen, fibronectin, and human thrombin. FG Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 3 of 11 was administered topically before wound closure using a surgical wound, hematoma management, surgical reop- special syringe supplied with the product. All researchers eration, and units of blood received. followed the instructions on the product data sheet. Unitary costs were obtained from the following data sources: a) the Financial Department of Hospital de la Outcomes Santa Creu and Sant Pau, Barcelona, Spain; b) the data- Results about efficacy and safety of assessed interventions base of the Spanish Hospital Costs Network (RECH, s.f.); were previously published [5]. Main outcome was blood and c) the scientific articles carried out in a similar con- loss collected postoperatively in the vacuum drains. Sec- text to our study [14, 15]. The cost of FG was obtained ondary outcomes were total blood loss, perioperative from a US study [16]. All costs were updated to the 2015- blood transfusions, units of transfused blood, postopera- year value and were reported in the value of the US Dol- tive complications, the length of hospital stay, treatment- lar adjusted by purchasing power parity (USPPP) («PPP. related adverse events, and mortality. Purchasing power converter», s.f.) (Table 1). Health‑related QoL Statistical analysis QoL was measured using the validated Spanish version of Statistical analysis of QoL the generic questionnaire EQ-5D-5L (EuroQoL) [11]. Statistical analyses were performed according to a modi- Before surgery, patients were asked about their QoL fied intention to treat approach, including all patients prior to the fracture, which was considered the base- who were randomized, received the surgical interven- line value. We also assessed QoL after the fracture but tion, and had assessments from at least two time points before surgery, at 5 days post-surgery, and at 1-, 6-, and including a pre-surgical and post-surgical QoL. Patients 12-months post-surgery. The patient interviews were with cognitive impairment were excluded from the analy- conducted in person during the hospitalization period sis because the QoL assessment could not be obtained. and by telephone after hospital discharge. When patients The latest available data were assigned to missing data were unable to answer the questions about QoL, a family points. member or caregiver responded. Differences in proportion between categorical data The EQ-5D-5L questionnaire assesses five dimensions: were analysed using Pearson’s Chi squared test. For con- mobility, personal care, carrying out daily activities, pain/ tinuous data, we calculated mean and standard devia- discomfort, and anxiety/depression. Each dimension has tion (SD) or mean difference and 95% confidence interval five severity levels (no problems, mild, moderate, severe, (CI), and differences were assessed using parametric or extreme problems, or disability) with an ordinal score t-test assuming a normal distribution. For multiple com- from 1 to 5. A score of 1 indicates a state of full health. parisons we used one way analysis of variance (ANOVA) In addition, this instrument performs a global assessment and the post-hoc Dunnett test. of the patient’s health through a vertical visual analogi- We evaluated QoL throughout the study, from baseline cal scale (EQ-VAS) that ranges from 0 to 100 (best score). to the end of follow-up. For changes in the VAS and EQ- The patient is questioned about how his/her health has 5D-5L scores, a two-way ANOVA was performed. The been on the day of the interview. factors were group (FG, TXA, and control), time (base- The partial results of the dimensions and the over- line, at day 5, and at 1, 6, and 12 months after surgery), all score of the EQ-5D-5L questionnaire were calcu- and interaction between them. The ANOVA was per- lated according to the scoring algorithms developed by formed using a general linear model (GLM) procedure. Ramos-Goñi et al. [12] QoL value for dead patients was Quality-adjusted life-years (QALYs) were calculated considered ‘zero’ from the date of death until the end of over the course of the 12 months using the area under follow-up at 12 months. the curve method based on the scores derived from the EQ-5D-5L questionnaire. Briefly, the QALY is a meas- Resources and costs ure of the value of health outcomes that combines the Total costs were calculated by multiplying the use of length of life and the quality of this life into a single value direct health care resources by their unit cost following (QALY) that can be compared across the three types of the ‘bottom-up’ costing method. The bottom up method treatments. To calculate the amount of time patients are estimates unit costs at the service provider level using alive in a state of health, we used the date from the col- detailed data from records [13]. lection of the EQ-5D-5L value, up to the end of follow-up We included drug-related direct costs (TXA and FG) or death. To calculate the quality of this life, we used the as well as medical care received during the hospitaliza- utility values derived from the EQ-5D-5L associated with tion period. Hospitalization costs included the hospital the state of health at day 5 and at 1, 6, and 12 months stay, medical complications, complications related to the after surgery. Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 4 of 11 Table 1 Baseline characteristics of patients included in QoL assessment Control FG Topical TXA p N = 46 N = 46 N = 42 n (%) n (%) n (%) Age (mean, SD) 84.06 (8.9) 81.74 (8.4) 83.57 (8.9) 0.412 Sex (women) 33 (71.7) 35 (76.1) 34 (81.0) 0.599 Medical background Hypertension 28 (60.9) 25 (54.3) 25 (59.5) 0.838 Surgery 19 (41.3) 20 (43.5) 20 (47.6) 0.493 Diabetes 11 (23.9) 11 (23.9) 5 (11.9) 0.287 Chronic obstructive pulmonary disease 2 (4.3) 8 (17.4) 5 (11.9) 0.128 Chronic renal disease 6 (13.0) 4 (8.7) 3 (7.1) 0.647 Digestive ulcer 1 (2.2) 2 (4.3) 0 - Haematological disease 0 2 (4.3) 0 - Chronic liver disease 2 (4.3) 0 0 - ASA I 1 (2.2) 1 (2.2) 0 - II 19 (41.3) 16 (34.8) 20 (47.6) III 18 (39.1) 9 (19.6) 13 (30.9) IV 1 (2.2) 2 (4.3) 0 Unknown 7 (15.2) 18 (39.1) 9 (21.5) Surgery duration (Minutes, mean and SD) 77.9(24.33) 79.3 (23.53) 80.3 (20.3) 0.616 Prosthesis Monopolar 24 (52.2) 19 (43.2) 15 (35.7) 0.289 Bipolar 20 (43.5) 19 (43.2) 22 (52.4) Total 2 (4.3) 6 (13,6) 5 (11.9) Type of prosthesis Austin Moore® 24 (52.2) 23 (50.0) 17 (40.5) 0.638 HA Karey® 15 (32.6) 14 (30.4) 18 (42.9) Other 7 (15.2) 9 (19.6) 7 (16.6) SD Standard Deviation, ASA Scale of surgical risk assessment of the American Society of Anaesthesiologists, TXA Tranexamic Acid, FG Fibrin Glue Death was assigned a utility of 0. The EQ-5D-5L utili- We first ranked the strategies by increasing cost. We ties varied between 0 (worst state of health) and 1 (per- used the definition of dominance provided by the York fect health status). To calculate the QALY, the two Health Economics Consortium: A dominant treatment measures are multiplied. For instance, one year lived option is one that is both less costly and results in better in perfect health required a utility value of 1 at each health outcomes than the comparator treatment (the for- time of evaluation (at day 5 and at 1, 6, and 12 months mer ‘dominates’ the latter). Conversely, a treatment option after surgery) and equates to 1 QALY. We summed the that is both more expensive and results in poorer health QALYs calculated at each of the evaluated times to obtain outcomes is said to be ‘dominated’ [17]. The strategy that the utilities gained by one patient in one year. We then was more costly and less effective than the preceding alter- summed all QALYs gained in each treatment group to native was ruled out by simple dominance. Then, we calcu- compare across treatment groups. lated the incremental cost-effectiveness ratio (ICER), which We used a two-sided significance level of 0.05. All is the ratio of the incremental cost difference ­(Cintervention statistical analyses were performed with IBM Corp. – ­Ccontrol) and the difference in QALYs ­(QALYintervention – Released 2017. IBM SPSS Statistics for Windows, Version ­QALYcontrol) obtained with one intervention with respect to 25.0. Armonk, NY: IBM Corp. the control: ICER = (Cintervention − Ccontrol )∕(QALYintervention − QALYcontrol ). Cost‑effectiveness analysis This economic evaluation was developed from the Span- A deterministic sensitivity analysis was carried out ish Health System perspective. The time horizon of this to assess the effect of the variation of both the utilities study was 12 months. No discount rate was applied. obtained with topical TXA as well as its cost. Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 5 of 11 All the analyses were performed using the statistical Health‑related QoL package Excel, Redmond, Washington (2011). The global assessment of QoL, measured using a VAS, was 68.6 at baseline (before fracture) with no statistically Results significant differences between groups (p = 0.517). After Characteristics of the study population the fracture and prior to surgery, the global QoL assess- One hundred and sixty-one patients were rand- ment was 30.9, with no statistically significant differences omized. Two patients died before the surgical inter- between groups (p = 0.512). vention, three patients withdrew their informed In successive follow-ups, there was a progressive improve- consent and in 21 patients the QoL could not be col- ment in QoL in all groups, but without reaching baseline lected due to significant cognitive limitations. The values. QoL at 12 months was 63.1 (18.8) for TXA, 58.0 number of exclusions was not statistically signifi- (19.9) for FG and 51.2 (26.9) for the control group (p = 0.045) cantly different between the groups (p = 0.23). Forty- (Table 3). VAS over the 12-month follow-up improved sig- six patients were included in the haemostasis group nificantly within each group (p < 0.001), but there were no (control), 42 in the topical TXA group and 46 in the differences between groups (p = 0.331) (Fig. 2). FG group (Fig. 1). The QoL index measured with the EQ-5D-5L instru- Baseline characteristics of the population included ment averaged 0.676 (0.272) at baseline (before fracture), for QoL assessment were similar between the study was negative before surgery (-0.383) and improved after groups (Table 1). There were also no significant dif- surgery, to 0.527 (0.379) at 12 months. There were no ferences in hospital stay or medical complications differences between groups in the QoL index (Table 3). (Table 2). During the 12-month follow-up, eight EQ-5D-5L over the 12-month follow-up improved sig- patients died in the control group, one in FG and one nificantly within each group (p < 0.001) but there were in TXA (p = 0.008). no differences between groups (p = 0.121) (Fig. 3). Fig. 1 Study flow chart of patients that undergoing Quality of Life Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 6 of 11 Table 2 Outcomes related with cost Control FG Topical TXA P n = 46 n = 46 n = 42 n (%) n (%) n (%) Hospital stays (mean days, SD) 11.3 (12.7) 11.3 (6.1) 11.6 (9.3) 0.987 Medical Complications Patients transfused 20 (43.5) 21 (45.7) 14 (33.3) > 0.461 Number of red blood transfusions 37 44 25 0.272 Wound infection 3 (6.5) 0 2 (4.8) 0.076 Urinary tract infection/UAR​ 2 (4.3) 6 (13.0) 1 (2.4) 0.100 Cardiac or respiratory failure/ Respiratory Infection 1 (2.2) 2 (4.3) 0 0.387 Prosthesis complications 3 (6.5) 3 (6.5) 0 0.238 Surgical reintervention 4 (8.7) 2 (4.3) 5 (11.9) 0.467 Hematoma 3 (6.5) 7 (15.2) 2 (4.8) 0.381 Mortality 8 (17.4) 1 (2.2) 1 (2.4) 0.007 TXA Tranexamic Acid, FG Fibrin Glue Table 3 QoL by haemostatic interventions of patients that underwent surgery after a femur fracture Control FG Topical TXA All patients p n = 46 n = 46 n = 42 N = 134 mean (SD) mean (SD) mean (SD) mean (SD) EQVAS Baseline 66.3 (16.2) 69.4 (18.4) 70.6 (13.6) 68.6 (16.2) 0.517 Pre-surgical 32.1 (11.0) 31.3 (14.2) 29.1 (11.8) 30.9 (12.6) 0.512 Post-surgical 41.8 (16.8) 43.9 (15.3) 43.4 12.9) 43.0 (15.1) 0.782 1 month 54.4 (21.7) 56.2 (16.6) 55.9 (15.9) 55.5 (18.2) 0.889 6 months 52.9 (24.6) 58.6 (15.9) 61.7 (15.5) 57.7 (19.0) 0.980 12 months 51.2 (26.9) 58.0 (19.9) 63.1 (18.8) 57.1 (22.6) 0.045 EQ5D5L index Baseline 0.634 (0.229) 0.689 (0.317) 0.714 (0.268) 0.676 (0.272) 0.456 Pre-surgical -0.390 (0.227) -0.317 (0.323) -0.447 (0.174) -0.383 (0.255) 0.054 Post-surgical 0.104 (0.277) 0.221 (0.244) 0.139 (0.286) 0.155 (0.271) 0.103 1 month 0.355 (0.353) 0.443 (0.297) 0.400 (0.351) 0.399 (0.333) 0.459 6 months 0.442 (0.374) 0.533 (0.366) 0.593 (0.312) 0.520 (0.356) 0.133 12 months 0.435 (0.383) 0.529 (0.417) 0.626 (0.306) 0.527 (0.379) 0.059 TXA Tranexamic Acid, FG Fibrin Glue, VAS Visual Analogical Scale Resources used and costs Cost‑effectiveness analysis We did not observe differences in hospital stay When comparing the cost-effectiveness of the interven- costs or in medical complications among groups tions, FG was ruled out by simple dominance since it (Table 4). However, the cost of one FG treatment was more costly than TXA (13,314.7 USPPP vs. 13,295.2 was significantly higher than the cost of the TXA USPPP, respectively), and less effective (0.0532 vs. or usual haemostasis (399.1, 12.9, and 0 USPPP, 0.0734, respectively). TXA compared to usual haemosta- respectively). sis had an ICER of 15,289.6 USPPP per QALY (Table 6). The average cost per patient in the haemostasis group was 12,886 USPPP, versus 13,351 USPPP in the topical TXA group, and 13,483 USPPP in the FG group. Both Deterministic sensitivity analysis the mean and median cost did not show significant dif- The deterministic sensitivity analysis was displayed ferences among the groups (Table 5). in a tornado diagram (Fig. 4). Regarding the variation Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 7 of 11 Fig. 2 Evolution of QoL related to health for global visual analogue scale and by treatment. Notes: Multi-factor ANOVA. VAS evolution at post-surgery 5th day, 1-, 6- and 12-months follow-up: p < 0.001. Comparison between treatments at post-surgery 5th day, 1-, 6- and 12-months follow-up: p = 0.331 Fig. 3 Evolution of QoL related to health for EQ5D5L and by treatment. Notes: Multi-factor ANOVA. EQ5D5L evolution at post-surgery 5th day, 1-, 6- and 12-months follow-up: p < 0.001. Comparison between treatments at post-surgery 5th day, 1-, 6- and 12-months follow-up: p = 0.121 in TXA cost, this ranged from zero (cost of usual hae- (0.040027) value of the 95% CI of the mean observed mostasis) to 25.8 USPPP, which is 100% higher than QALYs (0.073366). the unit cost of TXA. For the variation in the topical We found that the ICER was more sensitive to TXA QALYs, we used the upper (0.106704) and lower changes in TXA utilities than to its costs. An increase Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 8 of 11 Table 4 Unitary costs used in the study, updated to 2015-year value, and reported in $USPPP Source Published cost Country Publication year 2015-year value 2015 USPPP Treatments Control HSCSP 0.0 Spain 2015 0,0 0.0 Topical TXA HSCSP 8.6 € Spain 2015 8.6 € 12.9 FG Corral 2016 [16] 399.1 $US US 2016 399.1 $US 399.1 Hospital stays HSCSP 601.2 € Spain 2015 601.2 $ 916.5 Medical complications Blood transfusion (unit cost) Darbà 2009 [15] 350.0 € Spain 2009 395.3 $ 591.7 Wound infection RECH 3191.0 € Spain 2015 3191.0 $ 4776.9 Urinary tract infection RECH 1448.0 € Spain 2015 1448.0 $ 2167.7 Cardiac or respiratory failure Allué 2014 [14] 11,107.0 € Spain 2014 11,907.2 $ 17,825.1 Prostesis complications Allué 2014 [14] 7024.0 € Spain 2014 7530.0 $ 11,272.5 Surgical reintervention RECH 9579.0 € Spain 2015 9579.0 $ 14,339.8 Hematoma Allué 2014 [14] 3498.0 € Spain 2010 3750.0 $ 5613.8 HSCSP Financial Department of the Hospital de la Santa Creu and Sant Pau (Barcelona, Spain), RECH database of the Spanish Hospital Costs Network, USPPP $US Purchasing Power Parities are the rates of currency conversion that equalise the purchasing power of different currencies by eliminating the differences in price levels between countries, TXA Tranexamic Acid, FG Fibrin Glue, € Euros, $ Dollar Table 5 Average costs in $USPPP according to the intervention was 13,186 USPPP when the cost of TXA was assumed group to be zero. Control n = 46 FG n = 46 Topical TXA P n = 42 Discussion We have presented the results on QoL and compared Meana 12,703.8 13,314.7 13,295.2 0.965 the cost-effectiveness of TXA, FG, and surgical dia- SD 14,483.8 8,802.3 13,372.1 thermy for haemostasis in patients who underwent hip 95%CI 8,518.2 – 10,770.9 9,251.0–17,339.4 replacement surgery and agreed to participate in a ran- 16,889.4 -15.858,4 domized clinical trial that assessed the efficacy of these Medianb 9,431.8 10,671.4 10,094.3 0.188 three interventions to reduce perioperative blood loss Min 3,666.0 4,065.1 5,187.1 [5]. In our study, there were no significant differences in Max 100,756.8 52,993.7 86,483.0 QoL and costs between treatment groups. However, the Sum 584,374.0 612,474.3 558,399.6 analysis from the perspective of the payer showed that TXA Tranexamic Acid, FG Fibrin Glue, USPPP $US Purchasing Power Parities. FG proved to be a dominated intervention, being more a ANOVA test, bMedian difference test expensive and offering fewer QALYs than topical TXA. Furthermore, we found that the ICER was more sensi- in utilities to a mean value of 0.1067 decreased the ICER tive to changes in TXA utilities than its costs. to 7,660 USPPP, whereas when utilities decreased to a QoL improved after the surgical intervention but did mean value of 0.0400, the ICER increased to 56,136 not reach baseline pre-fracture values. These results are USPPP. In contrast, the effect of variation in TXA costs in line with other studies that demonstrated an improve- was less important. An increased cost of 25.8 USPPP ment, especially in the first 6 months after the injury, increased the ICER to 13,774 USPPP, whereas the ICER without reaching pre-fracture value [18–21]. In our Table 6 Cost-utility analysis of control, topical TXA and FG for prevention of post-surgical bleeding Treatment group Costs ($USPPP) Utilities (mean) Incremental cost Incremental effect ICUR​ ($USPPP) Control 12,703.8 0.029491 0.0 0,0 0 Topical TXA 13,295.2 0.073366 591.5 0.043875 13,480.3 FG 13,314.7 0.053200 19.4 -0.020200 Dominated $ USPPP $US Purchasing Power Parities are the rates of currency conversion that equalize the purchasing power of different currencies by eliminating the differences in price levels between countries, ICUR​ Incremental cost-utility ratio, TXA Tranexamic Acid, FG Fibrin Glue Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 9 of 11 Fig. 4 Sensitivity analysis of the incremental cost-utility ratio (ICUR). Notes: QALYs: Quality adjusted life years. The variation in QALYs was the 95% Confidence Interval upper (0.106704) and lower (0.040027) value of the mean utility (0.073366) in the topical TXA group. USPPP: $US Purchasing Power Parities are the rates of currency conversion that equalise the purchasing power of different currencies by eliminating the differences in price levels between countries. The variation in topical Tranexamic Acid (TXA) cost ranged from zero (control cost) and 25.8 USPPP, which is 100% higher than the unit cost of TXA study, QALYs were slightly higher in the group treated on unit cost did not come from a single source, which with topical TXA both at six and 12 months of follow- could affect the cost estimate. Another limitation was the up, but without reaching statistical significance. lack of a probabilistic sensitivity analysis, which reduces Patients in the TXA group had less perioperative bleed- the generalizability of the results to other contexts. The ing and transfusion requirements, and lower mortality, fact that we used the perspective of the payer and incor- which, although not statistically significant, together would porated only the costs associated with the care received contribute to a greater overall patient recovery. The ICER during hospitalization could also affect the accuracy of the calculated in our study was extremely sensitive to the prof- estimation of the resources used. However, given that the its obtained. In fact, in the sensitivity analysis, we observed study was randomized, it can be assumed that health care that a small variation, coinciding with the 95% CI values of was similar in the three compared groups, and in fact, this the average utility observed, can greatly affect the ICER. was verified with the similarity of baseline characteristics. Different studies have reported a reduction in direct The strengths of our study include the use of the care costs by decreasing costs associated with blood validated EQ-5D-5L instrument to measure QoL and transfusion, laboratory testing, and length of stay when QALYs. In addition, we have not identified other stud- TXA is used, compared with the standard care. These ies that evaluate the cost-effectiveness of treatments studies showed an average saving of $US 331 [22] or € to reduce post-surgical bleeding in our context. Some 284 [23] per patient when using TXA, which represents a authors have calculated the direct costs of the drug, 3-% reduction in hospital costs [22]. without considering the total cost of hospital medical Based on our results, FG does not seem to be a rec- care [24, 25]. One study analysed QoL and direct cost ommendable intervention from the point of view of effi- associated with the use of topical TXA in hip prosthesis ciency as it proved to be a dominated strategy, probably [26], reporting a better QoL index in the topical TXA due to its cost. group than with placebo (0.715 vs. 0.686 respectively) Limitations of our study were the open design with a but with follow-up limited to 3 months after surgery. potential reporting bias and the sample size was calculated Likewise, another study analysed direct cost of the use for the main outcomes of efficacy (blood loss), and not to of TXA in different presentations in patients with hip evaluate differences in QALYs. Secondly, the information fracture [27]. This study reported the lowest cost when Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 10 of 11 topical TXA was associated with intravenous TXA in Funding This project was funded by the Spanish Ministry of Health and Social Policy comparison with placebo, but only bleeding and throm- (grant code EC11-341), the European Regional Development Fund (FEDER) boembolic events were evaluated. and MAPFRE Foundation (grant code SA/12/AYU/456). In short, patients with a proximal femur fracture, QoL Availability of data and materials improved after surgery without reaching baseline values The datasets generated and/or analysed during the current study are not at 12-months of follow-up. There were no significant dif- publicly available due was not planned at the beginning of the study but are ferences in changes in QoL after surgery between the available from the corresponding author on reasonable request (Dr. Mª José Martínez-Zapata; e-mail: mmartinezz@santpau.cat). three interventions evaluated. Conclusion Declarations Topical TXA can be an efficient intervention in patients Ethical approval and consent to participate with femur fractures. The application of this drug repre- The protocol was approved by the clinical research ethics committees of the participating hospitals, as well as by the Spanish Agency for Medicines and sents a similar or marginally higher cost and the possi- Health Products. The Spanish ethics committees (EC) that approved the clini‑ bility of obtaining a slight advantage in terms of QALYs cal trial were EC of Hospital Universitari Germans Trias i Pujol de Badalona, EC when compared to usual haemostasis. FG would not be a of Hospital Universitari Mútua Terrassa, EC of Consorci Sanitari de Terrassa, EC of Fundació de Gestió Sanitaria Hospital de la Santa Creu i Sant Pau and, EC of first-choice treatment because it was a dominated strat- Hospital Clinic Barcelona, Spain. egy (i.e. more expensive and less effective). Future studies The study was conducted in accordance with the 1964 Declaration of Helsinki well-powered, multicentre, double-blind, or with assessor and its later amendments. Patients or their legal representatives gave signed informed consent to participate in the study before surgery. concealment and including patients reporting outcomes should confirm these results. Consent for publication Not applicable. Abbreviations Competing interests QoL: Quality of life; TXA: Tranexamic acid; FG: Fibrin glue; US: United States; $: The authors declare that they have no conflict of interest. Dollars; £: Pounds Sterling; RCT​: Randomized controlled clinical trial; USPPP: $US purchasing power parity; EQ-VAS: Visual analogical scale; SD: Standard Author details 1 deviation; CI: Confidence interval; ANOVA: Analysis of variance; QALYs: Quality- Iberoamerican Cochrane Centre‑Public Health and Clinical Epidemiology, adjusted life-years; ICER: Incremental cost-effectiveness ratio. IIBSant Pau, Barcelona, Spain. 2 Department of Social Medicine and Family Health, Universidad del Cauca, Popayán, Colombia. 3 Iberoamerican Cochrane Acknowledgements Center - Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain. 4 We would like to acknowledge Dr. Ignasi Gich for his support with the Department of Epidemiology and Evaluation, Hospital del Mar-IMIM, Barce‑ statistical analysis and Mrs. Anne Marshall for the English edition of the lona, Spain. 5 Iberoamerican Cochrane Centre-IIB Sant Pau, Barcelona, Spain. 6 manuscript. TRANEXFER Group: composed by the authors Julio De Caso, Orthopedic and Traumatology Service, Hospital de La Santa Creu I Sant Pau, Ion Carrera, Angie Millán, Mª Carmen Pulido, Marius Valera and Xavier Crusi: Barcelona, Spain. 7 CIBERESP, Hospital de La Santa Creu I Sant Pau, Sant Antoni from the Orthopedic and Traumatology Service, Hospital de la Santa Creu I María Claret 165, Pavilion 18, 08025 Barcelona, Spain. Sant Pau, Barcelona, Spain: José Antonio Fernández Núñez from the Anes‑ thesiology Service, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain; Received: 31 January 2022 Accepted: 16 August 2022 Anna Canalias Bage and Francesc Anglès Crespo from the Orthopedic and Traumatology Service, Hospital Universitari Mútua Terrassa, Barcelona, Spain; Anna Alavedra from the Orthopedic and Traumatology Service, Con‑ sorci Hospitalari Parc Taulí, Sabadell, Barcelona, Spain; Margarita Novellas from the Anesthesiology Service, Hospital Universitari Terrassa, Barcelona, References Spain; Gerard Urrutia from the Public Health and Clinical Epidemiology 1. Konda SR, Lott A, Manoli AI, Patel K, Egol KA. Hip Fracture Treatment at Service-Iberoamerican Cochrane Centre, IIB Sant Pau, CIBERESP, Barcelona, Orthopaedic Teaching Hospitals: Better Care at a Lower Cost. J Orthop Spain. Trauma. 2017;31:e364–8. 2. GBD 2019 Fracture Collaborators. Global, regional, and national burden of Authors’ contributions bone fractures in 204 countries and territories, 1990–2019: a systematic MJMZ, XA and MJ obtained funding for the projects. MJMZ, XA, MJMZ and EC analysis from the Global Burden of Disease Study 2019. Lancet Healthy obtained the data for the study. MP and AM analysed the data. AM, MP and Longev. 2021;2(9):580–92. MIMZ elaborated the first draft. All authors contributed in the final revision of 3. Sáez López P, González Montalvo JI, Ojeda Thies C, Gómez Campelo P. the study. All authors read and approved the final manuscript. Registro Nacional de Fracturas de Cadera (RNFC). Informe Anual 2018. http://​rnfc.​es/​wp-​conte​nt/​uploa​ds/​2020/​01/​Infor​me-​Anual-​RNFC-​ING-1.​ Authors’ information pdf (Accessed 12 July 2022). Dr. Angela Merchán is a PhD candidate in Methodology of Biomedical 4. Zhang P, He J, Fang Y, Chen P, Liang Y, Wang J. Efficacy and safety of Research and Public Health, Universitat Autònoma de Barcelona, Spain. She is intravenous tranexamic acid administration in patients undergoing hip funded by a predoctoral research contract (PFIS) from the Instituto de Salud fracture surgery for hemostasis: A meta-analysis. Medicine (Baltimore). Carlos III (FI17/00190). 2017;96(21):e6940. Dr. Mª José Martinez Zapata is funded by a Miguel Servet II research 5. Jordan M, Aguilera X, González JC, Castillón P, Salomó M, Hernández contract from the Instituto de Salud Carlos III (CPII20/00023). This project JA, et al. Prevention of postoperative bleeding in hip fractures treated was funded by the Spanish Ministry of Health and Social Policy (EC11-341) with prosthetic replacement: efficacy and safety of fibrin sealant and and from European Regional Development Fund (FEDER; “A way of making tranexamic acid. A randomised controlled clinical trial (TRANEXFER Europe”). study). Arch Orthop Trauma Surg. 2019;139:597–604. Merchán‑Galvis et al. BMC Musculoskeletal Disorders (2022) 23:827 Page 11 of 11 6. Mueller MM, Van Remoortel H, Meybohm P, Aranko K, Aubron C, Burger 26. Alshryda S, Mason J, Sarda P, Nargol A, Cooke N, Ahmad H, et al. Topical R, et al. Patient Blood Management: Recommendations From the 2018 (intra-articular) tranexamic acid reduces blood loss and transfusion rates Frankfurt Consensus Conference. JAMA. 2019;321:983–97. following total hip replacement: a randomized controlled trial (TRANX-H). 7. Papanicolas I, Figueroa JF, Schoenfeld AJ, et al. Differences in health JBJS. 2013;95:1969–74. care spending and utilization among older frail adults in high-income 27. Palija S, Bijeljac S, Manojlovic S, Jovicic Z, Jovanovic M, Cvijic P, et al. Effec‑ countries: ICCONIC hip fracture persona. Health Serv Res. 2021;56 Suppl tiveness of different doses and routes of administration of tranexamic 3(Suppl 3):1335–46. acid for total hip replacement. Int Orthop. 2020;45(4):865–70. 8. Williamson S, Landeiro F, McConnell T, et al. Costs of fragility hip fractures globally: a systematic review and meta-regression analysis. Osteoporos Int. 2017;28(10):2791–800. Publisher’s Note 9. Martínez-Zapata MJ, Jordán M, Aguilera X, Cánovas E, Urrutia G. Multi‑ Springer Nature remains neutral with regard to jurisdictional claims in pub‑ centre prospective cohort study to assess the quality of life of patients lished maps and institutional affiliations. undergoing surgery for subcapital femoral fracture. Trauma Fund MAP‑ FRE. 2014;25:188–95. 10. Husereau D, Drummond M, Petrou S, Carswell C, Moher D, Greenberg D, et al. Consolidated health economic evaluation reporting standards (CHEERS)—explanation and elaboration: a report of the ISPOR health economic evaluation publication guidelines good reporting practices task force. Value Health. 2013;16:231–50. 11. Hernandez G, Garin O, Pardo Y, Vilagut G, Pont À, Suárez M, et al. Validity of the EQ–5D–5L and reference norms for the Spanish population. Qual Life Res. 2018;27:2337–48. 12. Ramos-Goñi JM, Craig BM, Oppe M, Ramallo-Fariña Y, Pinto-Prades JL, Luo N, et al. Handling data quality issues to estimate the Spanish EQ-5D-5L value set using a hybrid interval regression approach. Value Health. 2018;21:596–604. 13. Chapko MK, Liu C-F, Perkins M, Li Y-F, Fortney JC, Maciejewski ML. Equiva‑ lence of two healthcare costing methods: bottom-up and top-down. Health Econ. 2009;18:1188–201. 14. Allué N, Chiarello P, Bernal Delgado E, Castells X, Giraldo P, Martínez N, et al. Impacto económico de los eventos adversos en los hospitales españoles a partir del Conjunto Mínimo Básico de Datos. Gac Sanit. 2014;28:48–54. 15 Darbá J, Restovic G, Arocho R. Coste de las transfusiones sanguíneas en España. Revisión de la literatura 2002–2007. PharmacoEconomics Span Res Artic. 2009;6:44–54. 16. Corral M, Ferko N, Hogan A, Hollmann SS, Gangoli G, Jamous N, Batiller J, Kocharian R. A hospital cost analysis of a fibrin sealant patch in soft tissue and hepatic surgical bleeding. Clinicoecon Outcomes Res. 2016;21(8):507–19. 17. Dominance - YHEC - York Health Economics Consortium (2016). https://​ yhec.​co.​uk/​gloss​ary/​domin​ance/ (Accessed 23 Dec 2021). 18. Brown K, Cameron ID, Keay L, Coxon K, Ivers R. Functioning and health- related quality of life following injury in older people: a systematic review. Inj Prev. 2017;23:403–11. 19. Campenfeldt P, Hedström M, Ekström W, Al-Ani AN. Good functional outcome but not regained health related quality of life in the majority of 20–69 years old patients with femoral neck fracture treated with internal fixation: A prospective 2-year follow-up study of 182 patients. Injury. 2017;48:2744–53. 20. Codesido P, Mejía A, Riego J, Ojeda-Thies C. Subtrochanteric fractures in elderly people treated with intramedullary fixation: quality of life and complications following open reduction and cerclage wiring versus closed reduction. Arch Orthop Trauma Surg. 2017;137:1077–85. 21. Alexiou KI, Roushias A, Varitimidis SE, Malizos KN. Quality of life and psy‑ chological consequences in elderly patients after a hip fracture: a review. Clin Interv Aging. 2018;13:143–50. Ready to submit your research ? Choose BMC and benefit from: 22. Cheung ZB, Anthony SG, Forsh DA, et al. Utilization, effectiveness, and safety of tranexamic acid use in hip fracture surgery: A population-based • fast, convenient online submission study. J Orthop. 2020;20:167–72. 23. Kyriakopoulos G, Oikonomou L, Panagopoulos A, Kotsarinis G, Vlachou • thorough peer review by experienced researchers in your field M, Anastopoulos G, et al. Transfusion rate, hospital stay and cost-effec‑ • rapid publication on acceptance tiveness of intravenous or local administration of tranexamic acid in total • support for research data, including large and complex data types hip and knee arthroplasty. A single-center randomized controlled clinical study. Orthop Rev. 2019;11:70–4. • gold Open Access which fosters wider collaboration and increased citations 24. Lopez-Picado A, Barrachina B, Remón M, Errea M. Cost-benefit analysis of • maximum visibility for your research: over 100M website views per year the use of tranexamic acid in total replacement hip surgery. J Clin Anesth. 2019;57:124–8. At BMC, research is always in progress. 25. Luo Z-Y, Wang H-Y, Wang D, Zhou K, Pei F-X, Zhou Z-K. Oral vs intravenous vs topical tranexamic acid in primary hip arthroplasty: a prospective, ran‑ Learn more biomedcentral.com/submissions domized, double-blind, controlled study. J Arthroplasty. 2018;33:786–93.