On the relevance of preoperative haemoglobin optimisation within a Patient Blood Management programme for elective hip arthroplasty surgery

Abstract

Background

Preoperative anaemia is highly prevalent among patients scheduled for total hip arthroplasty (THA), and is the main risk factor for perioperative red blood cell transfusion (RBCT). This retrospective cohort study aimed at assessing whether preoperative haemoglobin (Hb) optimisation reduced RBCT rates and improved outcome in this patient population.

Materials and methods

All patients entered a Patient Blood Management (PBM) programme consisting of in-hospital erythropoiesis stimulation, tranexamic acid administration, and a restrictive RBCT policy. Data from preoperatively anaemic patients (Hb <13 g/dL) who underwent THA, before (2015–2016, control group, n=75) or after (2017–2018, study group, n=70) the incorporation of a preoperative Hb optimisation protocol (Ferric carboxymaltose IV, 1,000 mg ± epoetin-α, 40,000 IU; administered 4 weeks prior to surgery) to the PBM programme underwent a comparative analysis.

Results

Haemoglobin concentrations at preoperative assessment were similar (12.1±0.7 g/dL vs 12.2±0.7 g/dL, for study and control groups, respectively; p=0.129). At hospital admission, significantly higher Hb were observed in the study group (13.4±0.8 g/dL vs 12.2±0.7 g/dL, respectively; p=0.001), with anaemia being corrected in 79% of cases. Compared to the control group, reduced perioperative RBCT rate (4% vs 24%, respectively; p=0.001), shorter length of hospital stay (6 [range 5–7] days vs 7 [5–8 days], respectively; p=0.002), and increased proportion of patients being discharged directly to their home (74% vs 47%, respectively; p=0.01) were observed in the study group. No treatment-related side-effects were witnessed.

Discussion

Within a PBM programme for THA, preoperative Hb optimisation was efficacious at correcting anaemia and minimising RBCT requirements, thus contributing to an improvement in postoperative outcomes.

INTRODUCTION

Total hip arthroplasty (THA) is considered a milestone in the history of orthopaedic surgery¹. It is a safe and successful surgical procedure, but is associated with significant perioperative bleeding. Although frequently under-estimated, the average blood loss in THA is around 1,000–1,500 mL^2,3. As a result, a significant proportion of THA patients may develop acute postoperative anaemia, which is usually treated by allogeneic red blood cell transfusion (RBCT). Nevertheless, RBTC rates after THA still largely depend on the transfusion thresholds of the individual institution^4,5.

Red blood cell transfusion is a costly and scarce resource. It is not without risk. Therefore the current trend is to restrict its use in order to prevent eventual complications (including postoperative infections and even mortality), as well as reduce long hospital stays and healthcare costs^6–9. In order to minimise RBCT in elective THA, a number of isolated blood saving measures were introduced in the 1990s. More recently, these measures have been integrated into a multimodal and multidisciplinary approach, called Patient Blood Management (PBM), which is based on the timely application of evidence-based medical and surgical concepts designed to stimulate erythropoiesis, optimise haemostasis and minimise blood loss in an effort to improve the outcome of patients at risk¹⁰. In addition, PBM implementation also results in improved transfusion practice and reduced healthcare costs.

In a previous randomised study, the addition of postoperative topical tranexamic acid (TXA) to our PBM programme for THA was shown to be efficacious in reducing both RBC mass loss and RBCT rate. Topical TXA was well tolerated, and no clinically apparent thromboembolic complications were observed. However, any beneficial effect of topical TXA on RBCT was restricted to THA patients with preoperative haemoglobin (Hb) ≥13 g/dL¹¹. This strongly suggested that its efficacy might be improved by reinforcing preoperative Hb optimisation^12,13.

Since January 2017, all THA patients presenting with Hb <13 g/dL at preoperative assessment with the anaesthesiologist and no contraindications were included in a preoperative Hb optimisation protocol with intravenous (IV) iron and recombinant human erythropoietin (rHuePO). This retrospective cohort study aimed at assessing whether the incorporation of preoperative Hb optimisation to our existing PBM programme reduced RBCT rates (primary objective) and improved postoperative outcome (secondary objective) in this patient population.

MATERIALS AND METHODS

Study design

We performed a retrospective analysis of the prospectively built database of consecutive elective THA carried out at the University Hospital “Miguel Servet”, during a 4-year period (January 2015 to December 2018). Data from anaemic patients (Hb <13 g/dL) who underwent THA surgery before the implementation of a preoperative Hb optimisation protocol within our PBM programme (2015–2016, control group, n=75) were compared to those of patients undergoing surgery after its implementation (2017–2018, study group, n=70). Only cemented or non-cemented primary THA were included. Patients were excluded if they presented with hip fracture, did not comply with the PBM programme for any reason, or if data records were incomplete. The study was reviewed, approved and registered by the Committee of Research Ethics of the Autonomous Community of Aragón, Spain (CEICA-C.I. PI 18/319).

Perioperative management

Surgery was scheduled during standard daytime working hours (Monday to Friday). All patients were operated on by the same surgical team, under standardised regional anaesthesia, antibiotic prophylaxis (cefazolin 2 g IV, 30 min before the intervention, or teicoplanin 600 mg in the case of drug allergy), surgical technique (posterolateral approach), choice of prosthetic implants (Anato stem with Trident PSL cup, Stryker Corporation, Kalamazoo, MI, USA; or cemented VerSys Heritage stem with ZCA cemented cup, Zimmer Inc., Warsaw, IN, USA), and postoperative analgesia. All received closed-suction drains, which were placed after wound closure and removed the next morning after the operation. All stayed in the post-anaesthesia recovery unit for at least four hours (h) before being transferred to the ward. Thromboprophylaxis was provide by once-daily, weight-adjusted dosing of low molecular heparin (Enoxaparin, Clexane, Sanofi-Aventis, Barcelona, Spain), which was started 12 h after surgery and maintained for the first 30 post-operative days.

Patient Blood Management programme

For the period January 2015-December 2016, the PBM programme in use at our centre consisted of in-hospital erythropoiesis stimulation, TXA administration, and a restrictive RBCT policy; this was applied to all study patients who underwent surgery during this period. For the period January 2017–December 2018, a preoperative Hb optimising protocol, aimed at achieving an Hb level >13 g/dL at hospital admission, was added to the existing PBM programme. This upgraded PBM programme was applied to all study patients who underwent surgery during this period.

Preoperative haemoglobin optimisation

Since January 2017, those presenting with Hb <13 g/dL at preoperative assessment with the anaesthesiologist and no contraindications received IV iron (1 g ferric carboxymaltose; Ferinject, Vifor, Saint Gallen, Switzerland) plus rHuEPO (40,000 IU, subcutaneous, sc; Eprex, Janssen-Cilag, Madrid, Spain) approximately four weeks prior to THA surgery.

In-hospital stimulation of erythropoiesis

Upon hospital admission, those without contraindications received: 1) IV iron (200 mg iron sucrose/48 h), three doses starting on admission; 2) parenteral vitamin B12 (1 mg); and oral folic acid (5 mg/day) for the entire duration of hospitalisation. Those with admission Hb concentration <13 g/dL also received an additional dose of rHuEPO (40,000 IU, sc) 24 h before surgery¹¹.

Tranexamic acid administration

Topical TXA (2 g, Amchafibrin, Meda Pharma S.L., Madrid, Spain) was administered following skin closure through the deeper drainage tube, which was subsequently clamped during the first 30 min after TXA dosing. At the anaesthesiologist’s discretion, some patients also received a single intraoperative IV dose of TXA (1 g).

Transfusion protocol

The anaesthesiologist estimated blood losses and made decisions on transfusion, both in the operating theatre and at the anaesthesia recovery unit. On the ward, the attending surgeon estimated postoperative blood loss and made decisions on postoperative transfusions. In these two groups of elderly patients, who may have a poor tolerance of anaemia, transfusion was indicated when the patient presented symptoms of acute anaemia (hypotension, tachycardia, tachypnoea, dizziness, fatigue, etc.), Hb level fell below 8 g/dL if there were no risk factors, or below 10 g/dL if there was ischaemic cardiomyopathy or severe peripheral vascular disease. The hospital blood transfusion committee approved this transfusion protocol, which was uniformly applied by anaesthesiologists and surgeons to all patients in the operating theatre, at the anaesthesia recovery unit, and on the ward for the entire duration of hospitalisation¹¹.

Data collection

A set of demographic and clinical data was gathered for all patients, including: gender, age, weight, height, body mass index (BMI), American Society of Anaesthesiologist (ASA) Physical Status Classification System scale, thromboembolic events and/or previous anticoagulant treatment, surgical procedure (duration, drainage output), perioperative Hb concentrations, RBCT rate (percentage of transfused patients) and index (RBCT units per patient), postoperative thromboembolic and infectious complications, adverse reaction to treatment (metallic taste, headache, nausea, vomiting, hypotension, anaphylactic reactions, seizures), length of hospital stay (LHS), and medical or prosthetic complications occurring within 30 days post surgery. Occurrence of deaths was followed-up for 6 months after surgery.

The Nadler’s formula¹⁴ was used to calculate the patients’ blood volume and the total RBC volume was derived by multiplying the blood volume with the corresponding haematocrit level. A factor of 0.91 was applied to correct haematocrit of peripheral blood sampling¹⁵. The overall perioperative RBC mass loss at postoperative day 1 was calculated by subtracting the RBC volume on postoperative day 1 from the preoperative RBC volume and by adding the total RBC volume transfused¹⁶ (1 RBC unit≈155 mL of RBC). To adjust baseline differences in total RBC volume, the lost RBC volume was also analysed as percentage of the patient’s baseline total RBC volume (relative RBC volume loss)¹⁶.

Statistical analysis

As this was a retrospective analysis, the study sample size was not calculated beforehand. Overall, 21 out of 148 patients received at least one RBCT during hospitalisation; of these, 18 out 75 (24%) were in the control group and 3 out of 70 (4%) in the study group. In a post-hoc calculation (http://www.gpower.hhu.de/en.html), the actual size of the study sample provided an alpha error of 2% and a statistical power of 95% for this RBCT reduction.

Data were expressed as number (percentages), mean±standard deviation (SD) or median [interquartile range]. Pearson’s χ² test or Fisher’s exact test was used for comparison of qualitative variables. Parametric two-way analyses of variance or non-parametric Kruskal-Wallis test were used for comparison of quantitative variables, after consideration of distributional characteristics. All statistics were performed with computer software (IBM SPSS 24.0, Chicago, IL, USA; licensed to the University of Málaga, Málaga, Spain). p<0.05 was considered significant.

RESULTS

A total of 713 patients underwent primary THA between January 2015 and December 2018. During data collection, 14 were excluded because of hip fracture surgery (n=9), non-compliance with the PBM programme (n=2), or incomplete records (n=3). Therefore, data from 699 were available for analysis (n=333 from the control group; n=366 from the study group) (Figure 1). Of those, 258 (77.5%) from the control group and 293 (80%) from the study group were excluded because they were non-anaemic (Hb ≥13 g/dL) at presentation with the anaesthesiologist (RBCT rates 1.6% and 2%, respectively). Three patients from the study group did not receive any preoperative Hb optimisation treatment, and were excluded from the analysis. Data from the remaining 145 (Hb <13 g/dL; control group, n=75; study group, n=70) were used for the comparative analysis (Figure 1).

Figure 1.

Distribution of patients scheduled for total hip arthroplasty (THA), according to group and haemoglobin concentration (Hb) at presentation with the anaesthesiologist for preoperative assessment

Control group: before the introduction of preoperative haemoglobin optimisation protocol. Study group: after the introduction of preoperative haemoglobin optimisation protocol. PBM: Patient Blood Management; N: number.

Demographic and clinical characteristics were similar in both THA groups, except for older age in the control group (p=0.027) (Table I). Similarly, there were no between-group differences in preoperative Hb or estimated blood volume, though preoperative RBC mass was slightly lower in the study group (Table II).

Table I.

Demographic and clinical characteristics of anaemic patients scheduled for primary total hip arthroplasty, before (control group) and after (study group) the implementation of a preoperative haemoglobin optimisation protocol

Characteristics	Control group	Study group	p
Patients	75	70	---
Gender (F/M)	60/15	56/14	1.000
Age (years)	71±12	66±16	0.027
BMI (kg/m2)	29±5	28±6	0.168
THA indication, n (%)			0.084
• Osteo-arthrosis	49 (65)	45 (64)
• Aseptic necrosis	24 (32)	16 (23)
• Hip dysplasia	1 (1)	8 (11)
• Other	1 (1)	1 (1)
ASA grade (I/II/III)	8/39/28	9/40/21	0.638
Co-morbidities, n (%)
• Hypertension	37 (49)	33 (47)	0.792
• Dyslipidaemia	27 (36)	18 (26)	0.181
• Diabetes	10 (13)	10 (14)	0.868
• Coronary disease	4 (5)	3 (4)	1.000
• Vascular disease	3 (4)	5 (7)	0.483
• COPD	9 (12)	7 (10)	0.794
• CKD	4 (5)	0 (0)	0.121
• Depression	3 (4)	9 (13)	0.071
Medication, n (%)			0.651
• Platelet anti-aggregant	9 (12)	8 (11)
• Oral anticoagulant	7 (9)	10 (14)

F: female; M: male; ASA: American Society of Anaesthesiologist Physical Status Classification System; BMI: body mass index; CKD: chronic kidney disease; COPD: chronic pulmonary obstructive disease; THA: total hip arthroplasty; n: number.

Table II.

Perioperative haemoglobin concentrations and transfusion requirements of anaemic patients scheduled for primary total hip arthroplasty, before (control group) and after (study group) the implementation of a preoperative haemoglobin optimisation protocol

Characteristics	Control group	Study group	p
Patients	75	70	---
EBV (L)	4.2±0.8	4.1±0.7	0.740
PreOP haemoglobin (g/dL)	12.2±0.7	12.0±0.7	0.129
PreOP RBC mass (mL)	1,568±286	1,485±244	0.064
Optimisation treatment			0.001
• FCM IV (1,000 mg), n (%)	0 (0)	70 (100)
• rHuEPO SC (40.000 U), n (%)	0 (0)	61 (87)*
HA haemoglobin (g/dL)	12.2±0.7	13.4±0.8	0.001
HA RBC mass (mL)	1,568±286	1,691±355	0.023
Anaemia correction n, (%)	0 (0)	55 (79)	0.001
PostOP haemoglobin (g/dL)	9.5±1.2	10.6±1.1	0.001
RBCT rate, n (%)	18 (24)	3 (4)	0.001
Total RBCT units	33	5	0.001
Haemoglobin post-RBCT (g/dL)	10.4±1.1	10.9±1.5	0.457

^*Nine patients from the study group did not receive rHuEPO due to contraindications (see Methods).

EBV: estimated blood volume; FCM: ferric carboxymaltose; HA: hospital admission; n: number; PreOP: preoperative; RBC: red blood cells; RBCT: packed red blood cell transfusion; rHuEPO: recombinant human erythropoietin.

Among the study group, 61 (87%) received FCM and rHuEPO, while 9 (13%) did not receive rHuEPO due to contraindications, and were treated with FCM monotherapy (Table II). Compared to the control group, the Hb optimisation treatment given to the study group resulted in higher Hb concentration (p=0.001) and RBC mass (p=0.023) at hospital admission, whereas anaemia was corrected (Hb ≥13 g/dL) in 79% of cases (Table II).

In the study group, mean increase in Hb was 1.3±0.9 g/dL (95% CI: 0.9–1.5 g/dL; p=0.001), and the RBC mass was increased by 205±160 mL (95% CI: 182–236 mL; p=0.001). Compared to the administration of FCM monotherapy, the combination treatment produced a higher increase in both Hb concentration (1.4±0.8 g/dL vs 0.7±0.7 g/dL, for FCM+rHuEPO and FCM, respectively; p=0.033) and RBC mass (211±160 mL vs 140±146 mL, respectively; p=0.207). However, no difference in anaemia correction rates were observed (6 of 9 [67%] vs 49 of 61 [80%], respectively; p=0.392), but it should be noted that Hb concentration at preoperative assessment was higher in the FCM subgroup compared to the FCM+rHuEPO sub-group (12.5±0.6 g/dL vs 12.0±0.7 g/dL, respectively; p=0.041).

There were no between-group differences regarding the use of cement and TXA, duration of surgery, 24 h postoperative drainage output, or perioperative RBC mass loss (Table III). Thus, at postoperative day 1, Hb concentration was significantly higher in the study group (p=0.001) (Table II), which translated into fewer patients requiring RBCT (3 [4%] vs 18 [24%], for study and control groups, respectively; p=0.001) and fewer packed RBC units transfused (5 units vs 33 units, respectively; p=0.001) (Table II). Most RBCT were given on the ward (72% vs 100%, respectively; p=0.579); only 5 patients from the control group received RBCT in the operating theatre and/or at the post-anaesthesia recovery unit. There were no differences in RBCT rates between those receiving FCM monotherapy (1 of 11; 9%) or FCM + rHuEPO (2 of 62; 3%) (p=0.549).

Table III.

Perioperative data and outcomes of anaemic patients scheduled for primary total hip arthroplasty, before (control group) and after (study group) the implementation of a preoperative haemoglobin optimisation protocol

Characteristics	Control group	Study group	p
Patients	75	70	---
Cemented, n (%)	58 (77)	46 (66)	0.121
Tranexamic acid, n (%)*			0.276
• Topical	48 (64)	40 (57)
• Topical + intravenous	27 (36)	28 (40)
Surgical time (min)	102±17	106±27	0.224
PostOP drainage output 24h (mL)	208±147	189±133	0.076
RBC mass loss (mL)	342±150	368±155	0.303
Relative RBC mass loss (%)	22±9	22±8	0.902
PostOP complications, n (%)**
• Medical	10 (13)	3 (4)	0.209
• Surgical	6 (8)	8 (11)	0.109
Length of stay (days)***	7 (5–8)	6 (5–7)	0.001
Length of stay >5 days, n (%)	56 (75)	36 (51)	0.004
Destination at discharge, n (%)			0.006
• Home	35 (47)	52 (74)
• Nursing home	18 (24)	8 (11)
• Family home	20 (27)	10 (14)
• Rehabilitation facility	2 (2)	0 (0)

^*Two patients from the study group did not receive tranexamic acid due to contraindication.

^**Medical complications: delirium (n=4), fever (n=2), urinary tract infection (n=2), paralytic Ileus (n=1), deep venous thrombosis (n=1), pulmonary thrombo-embolism (n=1), others (n=2). Surgical complications: wound disturbance (n=5), hip luxation (n=2), aseptic peri-prosthetic loosening (n=2), peri-prosthetic joint infection (n=2), peri-prosthetic fracture (n=3). One patient developed medical (delirium) and surgical (luxation) complications.

^***Median (interquartile range).

PostOP: postoperative; RBC: red blood cells; n: number; min: minutes; h: hours.

With regard to postoperative outcomes, there were no between-group differences in the incidence of medical or prosthetic complications (including thromboembolic and infectious complications) (21% vs 17%, for control and study groups, respectively; p=0.835). However, fewer patients from the study group stayed longer than 5 days in hospital (52% vs 75%, respectively; p=0.008) and more were discharged directly to their home (73% vs 43%, respectively; p=0.006) (Table III). There was no evidence of deaths in any group at 6 months of follow up.

DISCUSSION

Due to population ageing and improvements in healthcare standards, THA is an increasingly used procedure, with the number of interventions expected to have grown 174% by 2030¹⁷. However, THA surgery involves significant perioperative blood loss, which used to be treated by RBCT to avoid the deleterious effects of acute severe anaemia. The untoward effects of RBCT have led to the development of blood saving strategies, ultimately integrated into multimodal and multidisciplinary approaches, collectively called PBM programmes, aimed at improving patient outcome and avoiding unnecessary RBCT in primary THA^10,18.

In 1995, the RBCT rate in elective primary THA at our centre was 85%, which was progressively reduced to 32.6% with the introduction of blood saving strategies (in-hospital erythropoiesis stimulation, restrictive transfusion protocol)^19,20, and further decreased to 12% with the systematic use of topical TXA¹¹. However, the greatest benefit was observed in non-anaemic THA patients (preoperative Hb ≥13 g/dL)¹¹.

In THA surgery, preoperative anaemia increases by 4–6-fold the risk of perioperative RBCT^21,22. According to a recent International Consensus Statement, preoperative anaemia (or suboptimal preoperative Hb) should be defined as Hb <13 g/dL, irrespective of gender¹³. Using this definition, the prevalence of preoperative anaemia in our THA series was 21%, similar to that reported in other series²³ and in a systematic review²⁴. This high prevalence of preoperative anaemia has also been linked to greater postoperative morbidity and mortality, as well as longer LHS and higher healthcare costs^25–27. However, whether correction of preoperative anaemia can completely offset the high risk for postoperative complications other than those associated with RBCT, and which is the best approach to its pharmacological treatment are still a subject of discussion^12,13. In contrast, there is good evidence that correcting anaemia by RBCT can be detrimental to the outcomes of surgery²⁸.

Although the suggested benefits of PBM for both the patient and the health care system appear to be a matter of common sense, implementation is not necessarily straightforward (especially with regards to how to detect, diagnose and appropriately treat preoperative anaemia), and there are a number of obstacles that need to be overcome²⁸. A PBM programme for THA takes planning and forethought, as well as multidisciplinary collaboration^10,18,29.

In order to fulfil the three pillars of PBM, and with the synergistic work of the anaesthesia, orthopaedic, haematology and nursing departments, as of January 2017, our PBM programme for primary THA was upgraded to incorporate a preoperative Hb optimisation protocol for anaemic patients. The effectiveness of this strategy was assessed by comparing the RBCT rates (primary objective) and postoperative outcomes (secondary objective) in anaemic THA patients undergoing surgery before (2015–2016, control group) or after (2017–2018, study group) its implementation. A significant drop in RBCT rate was observed, from 24% in the control group to 4% in the study group (p=0.001). Despite the fact that patients from the study group were slightly younger than those from the control group (Table I), there were no between-group differences in the mean age of those receiving RBCT (72±14 years vs 69±17 years, for the control and study groups, respectively; p=0.731). Moreover, for the whole study sample, there were no differences in mean age between those transfused and those not transfused (72±14 years vs 69±14 years, respectively; p=0.352). However, there were no differences in the number of transfused units per transfused patient (1.8 U/patient vs 1.7 U/patient, respectively), indicating that adherence to current recommendations of transfusing one unit at a time, followed by post-transfusion evaluation, should be reinforced. Importantly, no treatment-related side effects were observed in our series.

In addition, shorter LHS (p=0.001) and an increased proportion of patients being discharged to their home (p=0.01) were observed in the study group, suggesting that higher postoperative Hb concentration and iron replenishment translated into improved physical fitness. However, this reduction in LHS is difficult to evaluate; in the absence of rigid criteria for discharge, it may be that standards had changed slightly during the study period. Of note, no between-group differences in postoperative complications were observed.

Regarding the treatment options, as absolute or functional iron deficiency is the most common cause of anaemia among orthopaedic surgical patients²³, iron supplementation should be implemented as early as possible before the scheduled procedure¹³. The National Institute for Health and Care Excellence in the UK (NICE) recommends offering oral iron before and after surgery to patients with iron deficiency anaemia³⁰. However, for patients not responding to oral iron, or intolerant to oral iron due to gastrointestinal side-effects, or with surgery planned less than 6 weeks in advance, IV iron may be the most effective option¹³.

Intravenous iron is highly efficacious at replenishing iron stores and increasing Hb in anaemia due to iron deficiency, with or without inflammation³¹. In elective THA, IV iron supplementation improves Hb levels and reduces RBCT requirements³². In practice, a dose of 1,000 mg is sufficient in most THA patients and, with the newer IV iron preparation (e.g. FCM or iron isomaltoside), it can be safely given by slow infusion over at least 15 minutes in one sitting³¹.

In Europe, rHuEPO administration is indicated to improve pre-operative Hb levels and reduce RBCT rates in patients undergoing elective orthopaedic surgery with moderate anaemia (Hb 10–13 g/dL) and expected to have moderate blood losses, in whom nutritional deficiencies have been ruled out, corrected or both³³. A recent metaanalysis of 5 randomised studies, including over 1,600 moderately anaemic THA and TKA patients, confirms that preoperative administration of rHuEPO significantly improves pre- and post-operative Hb and reduces RBCT rates, without increasing adverse side-effects³⁴. The use of erythropoietin (EPO) to improve preoperative Hb levels in orthopaedic surgery received also a conditional recommendation at the 2018 Frankfurt PBM Consensus Conference³⁵.

A significant proportion of anaemic orthopaedic surgical patients presented with laboratory parameters compatible with anaemia of chronic inflammation (with or without absolute iron deficiency)²³, thus further justifying the use of EPO¹³. However, the minimal effective rHuEPO dose to reach this goal is currently unknown, although studies in orthopaedic surgery patients demonstrated the superiority of the combined treatment with 1–2 doses of rHuEPO and iron over the single use of intravenous or oral iron therapy for optimising preoperative Hb concentration^36,37.

No differences in anaemia correction rates or RBCT rates were observed between FCM monotherapy and FCM+rHuEPO subgroups, but baseline Hb concentration was higher in the FCM monotherapy subgroup. In addition, the number of patients receiving FCM monotherapy was too small to allow any meaningful conclusion to be drawn. The retrospective nature of our single-centre study implies inherent constraints (risk of selection bias, risk of systematic errors in data, applicability) that limit its level of evidence. However, the inclusion of all anaemic THA cases that fulfilled the inclusion and exclusion criteria during the study period minimises the risk of selection bias. In the post-hoc analysis, the actual sample size provided an alpha error of 2% and a statistical power of 95% for the evaluation of the primary objective (RBCT rate). As a clinical ‘package’ promoting the implementation of a patient-centered and multimodal strategy, PBM is not amenable to randomisation because each patient is unique and therapeutic approaches will vary according to the clinical situation³⁸. In the present study, data were extracted from prospectively built clinical and surgical records, thus reflecting standard clinical practice and making any risk of systematic errors in data unlikely (no systematic overestimation of the magnitude of the effects of treatment)³⁸. Actual perioperative blood loss in orthopaedic surgery is frequently under-estimated, especially because of the significant proportion of hidden blood loss, which may even equal visible blood loss². In the Austrian benchmark studies, total blood loss was estimated at 3–4 postoperative day, as total RBC mass loss, taking into account the estimated blood volume, the preoperative and postoperative haematocrit, and the influence of RBC transfusions¹⁶. In the second Austrian benchmark study, a RBC loss ranging from 26% to 43% in THR procedures was observed, though there was a considerable inter-centre variability¹⁶. We use the same methodology to estimate % RBC mass loss in our patients (22% of RBC mass on the day prior to surgery, in both groups), but using the haematocrit value measured on the first postoperative day, which may be still influenced by the amount of fluid administered during surgery. As a consequence, total RBC loss might have been over-estimated and this should be regarded as another study limitation. Nevertheless, our estimated blood losses are in agreement with those recently published for a large multicentre cohort study of THA patients (n=4,662) who received TXA (IV, topical or combined)³⁹. Finally, this is a single-centre study, and though this ensured uniform patient care in both groups, the applicability of the study’s findings to other centres might be limited.

CONCLUSIONS

Despite the above-mentioned limitations, our data support the successful implementation of a preoperative Hb optimisation protocol within our PBM programme for patients scheduled for THA, and demonstrate its effectiveness for correcting preoperative anaemia and minimising RBCT requirements, thus contributing to improved postoperative outcome.