Perioperative management of the patients with hip fracture under anticoagulant or antiaggregants treatment. Consensus recommendations from the hemostasis section of SEDAR

Abstract

Background

Antiaggregant and anticoagulant therapy complicate the management of patients with osteoporotic hip fracture.

Objective

To homogenize and improve daily clinical practice with simple recommendations.

Methods

The haemostasis section of SEDAR established a working group to define an action plan for the management of antiaggregated or anticoagulated patients with an osteoporotic hip fracture. The suggested recommendations are based on evidence of best practices, and have been validated by a multidisciplinary group formed by 6 specialties.

Results

Early surgery reduces complications and mortality and improves patient comfort and functional recovery, with no difference in mortality between intradural and general anaesthesia.

Conclusion

Although uncertainties remain, it is recommended to perform surgery within 24−48 h of admission, adapting peripheral nerve blocks and type of anaesthesia (neuraxial or general) an to the haemostatic conditions. A multimodal management of antithrombotics, and the optimisation of haemostasis, haemoglobin and venous thromboprophylaxis since admission are suggested.

Resumen

Antecedentes

El tratamiento antiagregante y anticoagulante complica el manejo de los pacientes con fractura osteoporótica de cadera (FOC).

Métodos

La sección de hemostasia de la SEDAR constituyó un grupo de trabajo con el objetivo de establecer un plan de actuación basado en la evidencia, de las mejores prácticas en el manejo de los pacientes que estando anticoagulados o antiagregados sufren una FOC. Posteriormente un grupo multidisciplinar revalidó las recomendaciones.

Resultados

La cirugía precoz disminuye las complicaciones y la mortalidad, y mejora el confort y la recuperación funcional de los pacientes, mientras que la tasa de mortalidad no difiere por realizar una anestesia intradural o general.

Conclusiones

Aunque persisten incertidumbres, se recomienda realizar la cirugía en las 24−48 horas siguientes al ingreso, adaptando los bloqueos nerviosos periféricos y el tipo de anestesia (neuroaxial o general) a las condiciones hemostáticas. Se sugiere optimizar la hemostasia, la hemoglobina y la tromboprofilaxis desde el ingreso.

Introduction

Osteoporotic hip fracture (OHF), which has an Incidence of 40,000–50,000 patients/year, is a major problem in the Spanish National Health System.¹ It causes serious clinical complications, physical and cognitive dependence, is associated with a high mortality rate, and with a high consumption of health system resources.¹ Patients presenting OHF are typically frail, multi-morbid elderly patients (average age 87 years), most are women (76.5%), and they frequently present cognitive impairment or dementia (33%).¹ Between 40% and 50% of patients are treated with antithrombotics, either anticoagulants (20%–30%) or antiplatelet agents (70%-80%); acetylsalicylic acid (ASA) is the most prevalent therapy (60%).²

Antithrombotic therapy has a dual effect on OHF patients. On the one hand, surgical patients on antithrombotics have a higher risk of bleeding and intraoperative transfusion compared to patients not taking antithrombotics,³ and on the other hand, delaying surgery until its effect wears off does not reduce total transfusion requirements (from admission to discharge) and can promote complications and mortality.⁴ Delaying surgery by more than 48 h (versus < 48 h) can cause severe, persistent pain. This, combined with the frailty of the patients and the inflammation and hypercoagulability associated with OHF, increases the risk of delirium, cardiovascular events, venous thromboembolism, pressure ulcers, pneumonia, infections, dependency, and mortality.⁴ This dual effect has sparked debate on the optimal timing of surgery and other controversial issues, such as whether antithrombotics should be maintained, suspended, reversed, or replaced at the time of admission, whether they are a contraindication for peripheral nerve block (PNB), whether they are a contraindication for tranexamic acid (TXA), which anaesthesia technique should be used, and when antithrombotics can be restarted.

The goal of OHF treatment is to repair the fracture and return the patient to their previous level of physical and cognitive independence with the least amount of pain and shortest possible hospital stay.^5–8 The shorter the interval between admission and surgery and mobilization, the greater the functional recovery.^5–8 On the basis of these considerations, guidelines recommend performing surgery within 24–48 h of admission, irrespective of antithrombotic therapy.^5–8 In Spain, however, only around half of all OHF patients undergo surgery within 48 h (47.6% in 2022),^1,9 compared to more than 90% in neighbouring countries.^10,11 Prior treatment with antithrombotic drugs is one of the barriers to early surgery.^7,9 Firstly, because surgery is delayed in order to reduce the risk of bleeding and transfusion. Secondly, because the most commonly used anaesthesia technique in Spain is neuraxial anaesthesia (87%-89% of patients^1,10,11), and a pre-anaesthesia interruption interval needs to be observed in most antithrombotics.^12–14 This obviously delays surgery compared to other countries where most patients treated with antithrombotics receive general anaesthesia.^10,11 It is important to bear in mind that in patients treated with antithrombotics, delaying surgery by more than 48 h (compared to < 48 h) increases the risk of serious complications by around 20% and mortality at 1 month, 3 months and 1 year after OHF by 15%.^15–21 However, 1-, 3-, and 12-month mortality rates after OHF are similar in both neuraxial and general anaesthesia.^22–26 Each anaesthesia technique has particular advantages; general anaesthesia can be performed within 24−48 h, and neuraxial anaesthesia can reduce complications associated with difficult intubation and postoperative respiratory and kidney failure.^23–26 In addition, performing PNB at admission in patients with OHF can reduce the incidence of severe pain and preoperative delirium.^27–29 PNB combined with anaesthesia reduces local anaesthetic requirements, the incidence of neuraxial anaesthesia-induced hypotension, the need for intra- and postoperative opioids, and the incidence of severe pain, delirium and postoperative pneumonia.^27–29 In short, the type of anaesthesia used appears to have less impact on functional recovery, complications and mortality than early surgery. Therefore, notwithstanding exceptions, instead of delaying surgery until neuraxial anaesthesia is safe, it should be performed within the first 24−48 h using the most appropriate anaesthesia technique for the patient’s haemostatic status, the type of fracture, and the surgical approach.^{12–14,22–30}

In this context, the Spanish Hip Fracture Registry has called for the development of a specific perioperative management protocol that will improve care outcomes in OHF patients taking antithrombotics.⁷

Objectives

This document puts forward a multimodal perioperative strategy that involves performing PNB as soon as possible after OHF and performing surgery safely within 48 h of admission, despite antithrombotics.

What does this consensus bring to the table compared to existing guidelines and consensus?

Current recommendations on the perioperative management of patients on antithrombotic therapy undergoing scheduled and urgent surgery^12–14 do not consider the benefit of early PNB^27–29 + surgery within 48 h of admission, despite antithrombotics, in patients with OHF.^15–21 This means that there is less time than usually available in scheduled surgery to optimize haemostasis and haemoglobin levels, hence the need for an evidence-based management protocol for this specific population.^4–8

This document takes into consideration patient frailty, bleeding (associated with each type of OHF, surgical approach, and anaesthetic technique), the benefit of early multimodal analgesia with PNB, and the need to perform surgery within 48 h of admission. It also recommends antithrombotic management strategies, including optimising haemoglobin levels and haemostasis, administering venous thromboprophylaxis, and following PBM protocols from admission to discharge.

Material and method

The task force of the Haemostasis, Transfusion Medicine and Fluid Therapy Division of the Spanish Society of Anaesthesiology, Resuscitation and Pain Therapy (SEDAR) drafted the PICO questions (Patients, Intervention, Comparison and Results) shown in Table 1 to address the controversies currently surrounding the management of anticoagulated or antiplatelet patients with OHF. The searched Pubmed and the Cochrane library for the most relevant articles that answered the PICO questions in Spanish or English published between January 2015 and February 2024. The search terms were: “hip fracture” AND “time to surgery” “early mobilisation” “anticoagulation”, “INR”, “dicumarin” “direct oral anticoagulants” “antiplatelet“, “antiaggregant”, “aspirin”, “Clopidogrel”, “anaesthesia”, “nerve block”, “venous thromboembolism”, “thromboprophylaxis” “heparin low molecular weight” “iron” “tranexamic”, “transfusión”, “patient blood management”, “guidelines” “register”, and “meta análisis”. Fig. 1.

Table 1.

PICO questions on the management of osteoporotic hip fracture in patients taking anticoagulants or antiplatelet agents.

Supplement 1	Supplement 2	Supplement 3	Supplement 4	Supplement 5	Supplement 6	Supplement 7	Supplement 8
Delay and early vs. late mobilization in osteoporotic hip fracture (OHF) in patients taking antithrombotics	Neuraxial vs general anaesthesia	Peripheral nerve blocks (PNB) vs IV analgesia in patients taking antithrombotics	Patient blood management strategies in OHF + antithrombotics	Thromboprophylaxis in patients taking antithrombotics	Management of OHF in patients taking AVKs	Management of OHF in patients taking DOACs	Management of OHF in patients taking antiplatelet agents
1-Does early surgery (compared with delayed surgery) affect mortality and complications?	1-Is neuraxial anaesthesia (compared with general anaesthesia) associated with lower mortality and complication outcomes?	1-Is single-shot PNB at a compressible site (compared with IV analgesia) associated with a higher risk of complications?	1-Is Hb < 13 g/dL (compared with Hb > 13 g/dL) at admission associated with more complications, mortality, and hospital readmission?	1-Are different types of OHF and surgical approaches associated with different rates of DVT and pulmonary embolism?	1-In patients taking VKAs, does early surgery (compared with delayed surgery) affect mortality and complications?	1-In patients taking DOACs, does early surgery (compared with delayed surgery) affect mortality and complications?	1- In patients taking APA, does early surgery (compared with delayed surgery) affect mortality and complications?
2-Does early mobilization (compared with delayed mobilization) affect mortality, complications and discharge home?		2-Is PNB at a compressible site (compared with IV analgesia) associated with a higher risk of complications?	2-Are different types of OHF associated with different bleeding volumes and transfusion requirements?	2-Are age > 80 years (compared with age ≤ 80 years), frailty (compared with no frailty), and sex risk factors for DVT/PE?	2-In patients taking VKAs, does administering vitamin K in the emergency room facilitate early surgery?	2-In patients taking DOACs (compared with no antithrombotics), does early surgery affect 30-day complication or mortality rates?	2-In patients taking APA, does surgical delay affect bleeding and transfusion rates?
		3-Is single-shot PNB or continuous PNB at a non-compressible site (compared with IV analgesia) associated with a higher risk of complications?	3-In transfused (compared with non-transfused) patients, do ferritin, folic acid and vitamin B12 levels differ?	3-Can early surgery (compared with late surgery) reduce the incidence of DVT/PE?	3-In patients with OHF that are taking VKAs, does surgery with INR ≤ 1.5 (compared with INR ≤ 1.7) affect the transfusion rate?	3-In patients taking xabans with eGFR > 50 (compared with controls), does performing surgery within 24 h with ATX affect transfusion, thrombosis or 30-day mortality rates?	3-In patients taking APA and undergoing early surgery, does the type of antiplatelet agent affect bleeding volume and transfusion rates?
		4-In the preoperative period, is PNB at a compressible site more beneficial than administering IV analgesia alone?	4-Does treatment with APA (compared with no APA) increase bleeding after admission for OHF?	4-ln patients undergoing early surgery, does starting thromboprophylaxis before (compared with after) surgery affect rates of mortality, bleeding or DVT?	4-In patients taking VKAs and undergoing early surgery, does no preoperative LMWH (compared with preoperative LMWH) affect rates of transfusion, thrombosis or mortality?	4-In patients taking an non-reversed DOAC (compared with a reversed VKA), are there any differences in surgical delay, complications, readmissions, or mortality?	4- In patients taking APA (compared with controls), do post-OHF mortality and complication rates differ?
		5-Is the combination of PNB + spinal or general anaesthesia more beneficial than not using this combination?	5-In patients undergoing early OHF surgery, does treatment with antithrombotics (compared with no antithrombotics) increase the transfusion rate?	5-In patients taking antithrombotics, does the use of ATX (compared with no ATX) affect rates of mortality, DVT, PE, ACS or stroke in patients with high thrombotic risk (compared with no risk) or with ATX ≥ 20 mg/kg (compared with < 20 mg/kg)?	5-In severe VKA-related bleeding with INR > 2−2.5, is prothrombin complex concentrate (PCC) associated with better outcomes than fresh frozen plasma?	5-In patients undergoing early surgery, does taking xabans with eGFR < 30 (compared with eGFR > 30) or taking dabigatran with eGFR < 50 (compared with eGFR > 50) affect the transfusion rate?	5-Does performing PNB at a compressible site (compared with deep PNB) while maintaining APA or interrupting them for 24 h affect the risk of complications?
			6-Does combining antithrombotics with non-steroidal anti-inflammatory drugs or cox-2 inhibitors increase the Incidence of major bleeding?	6-Does the combination of IV iron and preoperative ATX + intraoperative topical ATX (compared with no ATX) affect the incidence of thrombosis?	6-In patients early surgery with INR ≤ 1.7, does taking VKAs (compared with no antithrombotics) affect the incidence of complications and mortality?	6- In patients taking a DOAC, is single-shot PNB at a compressible site associated with more risks than IV analgesia?	6-In patients taking clopidogrel ± ASA, how often is early OHF surgery performed under neuraxial (compared with general) anaesthesia?
			7-In patients taking antithrombotics, does the use of proton pump inhibitors (compared with no proton pump inhibitors) reduce the incidence of GI bleeding?	7-In patients with OHF, does thromboprophylaxis with LMWH (compared with other antithrombotics) reduce the rates of DVT, PE and mortality?	7-In patients taking VKAs and with INR ≤ 3, is PNB at a compressible site (compared with IV analgesia) associated with greater risks?	7-In patients undergoing early surgery that are taking a DOAC, does no preoperative LMWH (compared with preoperative LMWH) affect rates of mortality, major bleeding, transfusion, or thrombosis?	7-In patients taking clopidogrel ± ASA, what evidence is there for performing neuraxial anaesthesia after interrupting clopidogrel for up to 48 h?
			8-Does implementing a patient blood management (PBM) program (compared with no PBM) reduce transfusion requirements in patients with OHF?	8-In patients with OHF, can the use of mechanical devices reduce the incidence of DVT?	8-In patients taking VKAs and undergoing early surgery, is general anaesthesia more common than neuraxial anaesthesia?	8-In patients taking a DOAC, what is the evidence for performing neuraxial (compared with general) anaesthesia 48 h after admission?
			9-In patients with OHF, does a restrictive (compared with a liberal) transfusion strategy affect the incidence of mortality and serious complications?			9-Does restarting DOACs at a prophylactic dose 12−24 h after surgery (compared with prophylactic LMWH 12 h after surgery) affect the incidence of complications?
			10-In patients taking antithrombotics and undergoing and early surgery, does the use of TXA (compared with no ATX) affect the transfusion rate?
			11-Does IV (compared with topical) ATX affect plasma levels of ATX?
			12-Does IV iron (compared with no iron) affect HB, mortality, transfusion, infection, or hospital stay?
			13-Does the combination of IV iron + TXA + restrictive transfusion strategy affect the incidence of mortality and serious complications?

ACS: acute coronary syndrome; APA: antiplatelet agents; ASA: acetylsalicylic acid; DOAC: direct oral anticoagulants; DVT: Deep vein thrombosis; FFP: fresh frozen plasma; HB: haemoglobin; LMWH: low molecular weight heparin; OHF: osteoporotic hip fracture; PCC: prothrombin complex concentrate; PE: pulmonary embolism, PNB: peripheral nerve block; TXA: tranexamic acid; VKA: vitamin K antagonists. Early surgery: within 48 h of admission. Late surgery: 48 h after admission. Early mobilization: sitting within 72 h of admission.

Figure 1.

PRISMA flowchart of the literature review.

The GRADE (Grading of Recommendations, Assessment, Development and Evaluation) system was used to determine whether the level of evidence (LE) was high, moderate, low or very low. Each recommendation is based on the available evidence, and takes into account the benefit/risk balance and its impact on patients, anaesthesiologists, and the health system. A “strong recommendation” is given when the benefit clearly outweighs the possible adverse events, and a “weak recommendation” when the benefits and adverse effects are balanced or the evidence is of low quality. A “strong” recommendation is one that is fully endorsed, and a “weak” recommendation is merely a suggestion. The evidence found is summarised and presented in Supplements 1–8.

Three Delphi rounds were held in several stages to assign the level of each recommendation. The recommendations were then sent to a multidisciplinary panel of experts for their approval and comments.

Finally, the principle recommendations were summarised in an infographic (Fig. 2).

Figure 2.

Management of OHF in patient treated with antiplatelet agents or anticoagulants.

A. Types of fracture and surgical techniques. Adapted from: Fischer et al.,⁵¹ Zhu et al.,⁵² Vallon and Gamulin,⁵³ Rizkalla et al.⁵⁴

B. Stratification of thrombotic risk in patients treated with antiplatelet agents: Adapted from Sierra et al.¹⁴

C. Stratification of thrombotic risk in patients treated with anticoagulants: Prepared by the authors.

D. Pre- and postoperative management of antithrombotics in patients with hip fracture: Prepared by the authors.

Recommendations on the management of patients with an osteoporotic hip fracture (OHF) who are under treatment with antithrombotics

General recommendations for all antithrombotics: (Supplements 1–3) (infographic)

R-1 It is recommended to perform surgery within 24−48 h of admission, despite antithrombotic treatment, and to use this wait time to optimize haemostasis, haemoglobin and comorbidities (moderate LE, strong R).^{4–11,15–21}

R-2 It is recommended to start mobilizing the patient within 24 h of surgery, unless severely destabilised^4–11,21 (moderate LE, strong R).

R-3 It is recommended to adapt regional techniques (nerve blocks and neuraxial anaesthesia) to the patient’s haemostatic status, considering factors such as the type of antithrombotic drug, the time elapsed since the last dose, kidney function, and overall haemostasis^{4–14,16–20} (moderate LE, strong R). Deep PNBs require the same level of haemostasis as neuraxial anaesthesia, and PNBs at compressible sites do not require an antithrombotic interruption interval¹² (moderate LE, strong R).

R-4 It is recommended to evaluate the benefit of preoperative single-shot PNB in a compressible area (femoral, femoral triangle, fascia iliaca) as part of a multimodal analgesia regimen^27–29 (moderate LE, strong R).

R-5- It is recommended to perform surgery under a combination of PNB with sedation, general anaesthesia, or neuraxial anaesthesia (depending on haemostatic status) (moderate LE, strong R).^{5–9,27–29}

R-6 It is recommended to weigh up the risks and benefits of performing neuraxial anaesthesia with a shorter antithrombotic interruption interval than recommended for scheduled surgery against performing general anaesthesia combined with PNB in each individual patient^12–14 (moderate LE, strong R).

R-7 It is recommended to consider including postoperative PNB in a multimodal analgesia regimen if not already administered for surgery (moderate LE, strong R).^{5–9,27–29}

R-8 It is recommended to monitor for signs of spinal haematoma (low back pain, incontinence, and numbness or weakness of the lower extremities) or perineural haematoma (excessive duration of sensory and/or motor blockade) from the time regional anaesthesia is administered until after the introduction of thromboprophylaxis and antithrombotics¹¹ (moderate LE, strong R).

Optimizing haemostasis, anaemia and transfusion requirements (Supplement 4) (Infographic)

R-9 It is recommended to determine haemoglobin levels at admission (moderate LE, strong R)

R-10 It is suggested to regularly check haemoglobin levels, preferably using a digital meter, and to monitor for clinical signs of anaemia from admission until the fifth postoperative day, bearing in mind that haemoglobin levels will fall by between 2 and 6 g/dL (equivalent to losing 400 to 1200 mg of iron) during this time, depending on the type of fracture, the surgical technique used and the antithrombotic treatment administered³ (moderate LE, strong R).

R-11 It is recommended to follow patient blood management (PBM) protocols to optimize haemoglobin levels from the time of admission until after discharge^3,31–36 (moderate LE, strong R).

R-12 It is recommended to interrupt antithrombotics at admission, replacing antiplatelet agents with ASA 100 mg/24 h (or Triflusal 300 mg/24 h in patients with sensitivity to ASA),^12,13 avoid non-steroidal anti-inflammatory drugs, and administer gastrointestinal bleeding prophylaxis (moderate LE, strong R).

R-13 In hip fractures (except non-displaced subcapital fractures) diagnosed less than 3 h after the fall and scheduled for repair ≤ 48 h, it is suggested to consider administering tranexamic acid (TXA) (15 mg/kg or 1 g IV)³⁴ (low LE, weak R).

R-14 It is recommended to administer TXA before surgical incision (10−15 mg/kg or 1 g IV) in a single dose or combined with topical TXA (1.5−2 g) without exceeding a total of 20 mg/kg³⁵ (moderate LE, strong R).

R-15 It is suggested to consider reducing the IV dose of TXA (10 mg/kg) in patients with kidney failure, surgical delay > 72 h, hypercoagulability syndrome, previous thromboembolic episodes, a thrombotic event, or a stent with incomplete endothelialisation (conventional stent: <1 month, drug-eluting stent: <6 months) or in patients that received TXA in the emergency room^35,36 (moderate LE, weak R).

R-16 A postoperative haemoglobin transfusion threshold of 8–8.5 g/dL is recommended in asymptomatic patients, individualizing this threshold if bleeding persists, the patient presents ischaemic heart disease, is very elderly, or is very frail³³ (moderate LE, strong R).

R-17 It is recommended to restart antithrombotic treatment when postoperative bleeding is controlled and haemostasis has normalised^11–13 (moderate LE, strong R).

Venous thromboprophylaxis (Supplement 5) (infographic)

R-18 It is recommended to bear in mind that patients with OHF are at high risk of venous thromboembolic disease and bleeding, and individual risk factors should also be taken into consideration^3,37–40 (moderate LE, strong R).

R-19 It is recommended to standardise physical thromboprophylaxis measures in all patients, including: delay surgery ≤ 48 h, place the patient in a sitting position before surgery, start mobilization within 24 h of surgery, and use mechanical thromboprophylaxis^37,41 (moderate LE, strong R).

R-20 It is suggested not to administer preoperative low molecular weight heparin (LMWH) if surgery is delayed ≤ 48 h in patients taking anticoagulants or dual antiplatelet therapy⁴⁰ (low LE, weak R).

R-21 If surgical delay of >48 h is expected and bleeding is controlled, it is suggested to consider starting preoperative pharmacological venous thromboprophylaxis 12−48 h after admission^37,40,41 (low LE, weak R)

• -

  In patients taking anticoagulants: (VKA or DOAC): Prophylactic LMWH + physical thromboprophylaxis.

• -

  In patients taking antiplatelet agents: ASA 100 mg/24 h + prophylactic LMWH + physical thromboprophylaxis.

• -

  In patients with uncontrolled bleeding: physical thromboprophylaxis only.

R-22 If LMWH is administered, it is suggested to reduce the dose in patients with estimated glomerular filtration rate (eGFR) ≤ 30 mL/min/1.73m², underweight (≤50 kg), or with previous dual antiplatelet therapy^12,13 (low LE, weak R).

R-23 It is recommended to interrupt LMWH before surgery on the basis of factors such as eGFR and LMWH dose^12,13 (low LE, strong R).

• •

  eGFR > 30 mL/min/1.73m²

  • -

    Prophylactic dose → ≥ 12 h

  • -

    Intermediate dose or anticoagulant → ≥ 24 h

• •

  eGFR< 30 mL/min/1.73m²

  • -

    Prophylactic dose → ≥ 24 h

  • -

    Intermediate dose or anticoagulant → ≥ 48 h

R-24 It is suggested to start postoperative prophylactic dose LMWH 8−12 h after surgery (24 h after the previous LMWH, if administered), or to wait for 24 h after surgery if bleeding persists or in the event of traumatic of bloody intradural puncture^11–13,37 (low LE, weak R).

Antivitamin K (AVK) anticoagulants: acenocoumarol or warfarin (Supplement 6) (Infographic)

Preoperative period

R-25 It is recommended to stop AVK and administer vitamin K (10 mg IV) at admission (even in patients with high thromboembolic risk), perform INR at 8−12 h, and if INR > 1.5 → repeat vitamin K, and perform surgery within 24−48 h after admission; if INR ≤ 1.7 there is no need to delay surgery ^13,18 (moderate LE, strong R).

R-26 In patients with high thromboembolic risk, it is recommended to perform surgery within 24−48 h if INR ≤ 1.7 instead of using bridging therapy with LMWH or attempting INR ≤ 1.25^18,44 (low LE, strong R).

R-27 If surgery is expected to be performed > 48 h, it is suggested to use preoperative LMWH starting with INR < 1.8–2.

• •

  If the thromboembolic risk is not high: LMWH at prophylactic dose.^5,6,44

• •

  If the thromboembolic risk is high (mechanical valves or atrial fibrillation with thromboembolism in the preceding month): consider intermediate-dose LMWH or anticoagulant once a day to facilitate surgical scheduling^5,6,44 (low LE, weak R).

Surgery and anaesthesia

R-28 It is suggested to adapt the type of anaesthesia to the INR obtained immediately before anaesthesia and surgery^5,6,1–12:

• •

  INR ≤ 1.25: suitable for neuraxial or general anaesthesia (moderate LE, strong R).

• •

  INR 1.26 to 1.5: neuraxial anaesthesia or deep PNB is not contraindicated (low LE, weak R).

• •

  INR > 1.51−1.7: it is suggested to proceed with superficial PNB and general anaesthesia (low LE, weak R).

R-29 If intraoperative bleeding is excessive, it is suggested to monitor INR, and if it is >2−2.5 consider administering 8–10 IU/kg prothrombin complex concentrate^8,44 (low LE, weak R).

Postoperative period

R-30 If bleeding is controlled, it is suggested to start prophylactic LMWH 8−12 h after surgery (24 h after the previous LMWH, if administered) and to start VKA 24−72 h later (avoiding simultaneous administration of VKA with antibiotics, because they could interact and increase the risk of bleeding). It is suggested to administer LMWH and VKA simultaneously until INR ≥ 2, at which point LMWH should be discontinued^12,13,18,44 (low LE, weak R).

Direct oral anticoagulants (DOAC) (Supplement 7) (Infographic)

Preoperative period

R-31 It is suggested to monitor DOAC levels in patients with kidney failure (eGFR < 50 with dabigatran or eGFR < 30 with xabans), if available.^{4–6,12,13,18–20} It is suggested not to delay surgery due to alterations in standard coagulation tests, and if there are any, perform superficial PNB and general anesthesia.^{4–6,12,13,18–20,45–48} (low LE, weak R)

R-32 It is suggested not to reverse DOACs with specific reversal agents such as idarucizumab (available in Spain) or andexanet ᾱ (not currently available). It is also not recommended to administer prothrombin complex^{4–6,12,13,18–20,45–47} (low LE, weak R).

R-33 It is recommended not to use preoperative LMWH in patients treated with DOACs who undergo surgery within 48 h of admission^37,41,44 (low LE, strong R).

R-34 If surgical delay > 48 h is expected, it is suggested to administer preoperative prophylactic LMWH (low LE, weak R).

• •

  eGFR > 50, start prophylactic LMWH 24 h after the last dose of DOAC

• •

  eGFR < 50, start prophylactic LMWH 24−48 h after the last dose of DOAC

Surgery and anaesthesia

R-35 It is suggested to perform surgery within 24−48 h of the last dose of DOAC, adapting the type of anaesthesia to plasma DOAC levels, and if not available, to the type of DOAC administered, eGFR, and the time since the last dose^{3,12,13,45–48} (low LE, weak R). With normal fibrinogen levels and platelet count, the following is suggested^{3,11,12,45–48}:

With xabans (apixaban, edoxaban, rivaroxaban)

• •

  Plasma levels ≤ 30 ng/mL → suitable for neuraxial anaesthesia and deep PNB (moderate LE, strong R).

• •

  eGFR < 30 mL/min/1.73 m² (low LE, weak R):

  • -

    48 h: general anaesthesia and PNB at a compressible site.

  • -

    If general anaesthesia is contraindicated, surgery may be exceptionally delayed for 72 h after the last dose in order perform neuraxial anaesthesia.

• •

  eGFR≥ 30 mL/min/1.73 m² (low LE, weak R):

  • -

    24 h: general anaesthesia and PNB at a compressible site

  • -

    At 48 h after the last dose, intradural anaesthesia and deep PNB can be considered.

With dabigatran

• •

  Plasma levels < 30 ng/mL → suitable for neuraxial anaesthesia and deep PNB (moderate LE, strong R).

• •

  eGFR < 50 mL/min/1.73 m² (low LE, weak R):

  • -

    48 h: general anaesthesia and PNB at a compressible site.

• •

  eGFR≥ 50 and < 79 mL/min/1.73 m² (low LE, weak R):

  • -

    36 h: general anaesthesia and PNB at a compressible site.

  • -

    If general anaesthesia is contraindicated, surgery may be exceptionally delayed for 72 h after the last dose in order perform neuraxial anaesthesia.

• •

  eGFR≥ 80 mL/min/1.73 m² (low LE, weak R):

  • -

    24 h: general anaesthesia and PNB at a compressible site.

  • -

    At 48 h after the last dose, intradural anaesthesia and deep PNB can be considered.

Postoperative period

R-36 It is suggested to perform postoperative venous thromboprophylaxis using either LMWH or DOAC at prophylactic doses. Either should be started after considering plasma DOAC levels, and if unavailable, the surgical delay and eGFR (without simultaneous LMWH and DOAC on the same day)^{12,20,41,43–48} (low LE, weak R):

• •

  Surgical delay of 24 h: → start LMWH or DOAC 24 h after surgery.

• •

  Surgical delay of 36−48 h: → Start LMWH or DOAC 8−12 h after surgery.

• •

  If eGFR ≤ 30 mL/min/1.73m² it is suggested to use LMWH titrated to kidney failure.

R-37 At 48 h after surgery, it is suggested to switch to the usual dose of DOAC adapted to the eGFR,^{12,20,41,43–48} discontinuing LMWH if it was being administered (low NE, weak R).

Antiplatelet agents (APA) (Supplements 4, 5 and 8) (infographic)

Preoperative period

R-38 It is recommended to perform surgery within 24−48 h of admission in patients on any antiplatelet regimen. No delay is required in patients taking ≤200 mg ASA or ≤300 mg triflusal (moderate LE, strong R).

R-39 It is suggested to stop APA at admission and replace it with 100 mg ASA or 300 mg triflusal (in patients with sensitivity to ASA)^13,14 (moderate LE, weak R).

R-40 It is suggested to adapt the start of preoperative venous thromboprophylaxis with LMWH to the surgical delay and the bleeding risk associated with the antiplatelet regimen (moderate LE, weak R):

• •

  Clopidogrel/prasugrel/ticagrelor + ASA; it is suggested to start LMWH only if a delay of >48 h is expected, administering half the usual prophylactic dose from 24 h after admission if bleeding is controlled.^37,41,42

• •

  Other antiplatelet regimen:

  • -

    surgery ≤ 24 h: it is suggested not to use preoperative LMWH,^37,41,42

  • -

    surgery > 24 h: it is suggested to start administering 100 mg ASA + prophylactic LMWH 12−24 h after admission.

Surgery and anaesthesia

R-41 In patients taking ASA (≤200 mg) or triflusal (≤300 mg), it is recommended to administer neuraxial anaesthesia and any single-shot PNB¹⁴ (low LE, strong R).

R-42 It is suggested to perform a combination of single-shot PNB at a compressible site and general anaesthesia 24−48 h after the last dose of any APA (other than ASA ≤200 mg or triflusal ≤300 mg) in monotherapy or dual therapy^12–14 (low LE, weak R).

R-43 Exceptionally, if general anaesthesia involves a serious risk, it is suggested to consider neuraxial anaesthesia with a shorter antithrombotic interruption interval than recommended for scheduled surgery, only if the following 3requirements are met:^13,14,30,49 (low LE, weak R):

• •

  Normal platelet and fibrinogen count.

• •

  No haematologic disorder or need for anticoagulant therapy, and the recommended LMWH interruption interval has been met.

• •

  Estimated 70,000–80,000 functional platelets/µL are available¹³ (minimum necessary to perform neuraxial anaesthesia) in 1 of following situations:

  • 1

    Platelets aggregate on platelet aggregation tests or

  • 2

    After discontinuing APA and replacing it with 100 mg ASA, 10%–15% of inhibited platelets are renewed each day.¹⁴ Depending on the reversibility and potency of the APA, ≥70,000–80,000 functional platelets/µL would be achieved in the following time frames:

    • -

      24 h after the last dose of dipyridamole or cilostazol.

    • -

      48−72 h after the last dose of clopidogrel monotherapy.

    • -

      72 h after the last dose of clopidogrel/ticagrelor + ASA.

• •

  In patients taking prasugrel, it is suggested to perform PNB at a compressible site and general anaesthesia within 24−48 h of admission (low LE, weak R).

R-44 If intraoperative bleeding is excessive despite TXA, it is suggested to monitor haemostasis and consider platelet transfusion and/or administration of desmopressin (≤0.3 µg/kg)¹⁴ (low LE, weak R).

Postoperative period

R-45 If bleeding is controlled, it is suggested to restart antiplatelet agents, taking into account the interruption interval, the time to onset of action, and the potency of the APA^13,14,50 (low LE, weak R).

• •

  100 mg ASA and 300 mg triflusal: →6−12 h after surgery.

• •

  Clopidogrel, dipyridamole, cilostazol: →24−48 h after surgery.

• •

  Prasugrel, ticagrelor: →24−72 h after surgery.

R-46 It is suggested to prioritize restarting the patient’s chronic ASA and APA regimen together with physical venous thromboprophylaxis strategies over the introduction of LMWH (low LE, weak R).

R-47 It is suggested to perform postoperative venous thromboprophylaxis with prophylactic LMWH 8−12 h after surgery (24 h after the previous LMWH dose, if administered). In patients on clopidogrel/prasugrel/ticagrelor + ASA, it is suggested to administer half the usual prophylactic dose of LMWH (low LE, weak R).

R-48 In patients with high bleeding risk and a surgical delay of no more than 48 h, who are walking after hip fracture and do not have additional risk factors for deep vein thrombosis (DVT), after 14 days of treatment with LMWH it is suggested to consider continuing with LMWH or with one of the following thromboprophylaxis options:

• •

  Clopidogrel/prasugrel/ticagrelor + ASA + LMWH at half the usual prophylactic dose. If they are at high risk of bleeding, it is suggested to consider discontinuing LMWH if all 3 criteria are met^37,41,42 (low LE, weak R).

• •

  ASA 100 mg/24 h + LMWH. If they are at high risk of bleeding, it is suggested to consider performing venous thromboprophylaxis with ASA 50−100 mg/12 h and suspending LMWH if all 3 criteria are met^37,41,42 (low LE, weak R).

Discussion

The Haemostasis, Fluid Therapy and Transfusion Division of the Spanish Society of Anaesthesia, Resuscitation and Pain Therapy (SEDAR) has prepared 48 recommendations that together form a multimodal perioperative action plan for safe OHF surgery within 48 h of admission in patients receiving anticoagulation or antiplatelet therapy. These simple, practical recommendations optimize the perioperative management of these patients and can be easily implemented by a multidisciplinary team. They must be adapted to the resources available in each hospital after the attending team has performed a detailed analysis of the risks and benefits of delaying surgery, the best anaesthesia technique, surgical approach, and antithrombotic management strategy (including venous thromboprophylaxis), and the optimal strategy for optimising haemostasis and haemoglobin levels.

In patients treated with antithrombotics and with insufficient hemostatic conditions for an intradural technique in the first 24-48 h, it is considered optimal to perform superficial PNB and general anesthesia in the first 24-48 h. However, even if a particular patient’s haemostatic status does not fulfil the criteria usually required for scheduled surgery, neuraxial anaesthesia would be justified if it will improve morbidity and mortality (Supplements 6–8). This decision should be taken by a multidisciplinary team after performing a risk-benefit analysis and should be discussed with the patient or their caregiver. The balance of risks and benefits and the best balance of the anesthetic technique used should be reflected in the patient's medical record.

Clinicians should be aware of the recent paradigm shift in the approach to perioperative thromboprophylaxis.^37,41 The residual effect of VKA and DOACs appears to provide sufficient protection against thrombosis if surgery is performed within 48 h. This eliminates the need for preoperative LMWH, which in turn increases the safety of early surgery and regional anaesthesia. Furthermore, restarting DOACs at a prophylactic dose for venous thromboprophylaxis after interrupting them before surgery eliminates the need for LMWH, and has been shown to be good practice. ASA can reduce platelet aggregation and activation and prothrombin and fibrinogen activity while appearing to promote fibrinolysis.^40,42 ASA combined with early ambulation has been shown to protect against venous thromboprophylaxis in OHF, may cause less bleeding than other LMWH + antiplatelet agent combinations, is more comfortable for the patient, and is less expensive than other options;^37,40,41 however, there is far more evidence on the efficacy of LMWH for thromboprophylaxis in OHF than ASA. Once thromboprophylaxis with ASA 100 mg/12 h has been completed, ASA 100 mg/24 h should be maintained; antiplatelet therapy should never be permanently discontinued in patients that need it.

Supplements 1–8 summarize the available evidence on the management of OHF in patients treated with antithrombotics and the rationale behind the recommendations. The main limitation of the recommendations is that they are predominantly based on observational studies, and are therefore weak. Strong recommendations are based on randomised studies with small patient samples or on unbiased observational studies. Because of this, uncertainties persist, and the recommendations should be reviewed and modified as new evidence becomes available. Despite these recommendations, clinicians still need to analyse the particular circumstances of each patient in order to establish the most appropriate indications in each case and context. Follow-up studies on the proposed changes should be performed and protocols must be updated with new evidence.

Conclusions

The following recommendations for the management of OHF in patients treated with antithrombotics have been made: administer multimodal analgesia that includes a PNB at a compressible site at the time of admission; perform surgery within 48 h of admission under PNB and general or neuraxial anaesthesia, depending on the patient’s haemostatic status; follow patient blood management protocols from the time of admission (optimize haemoglobin and haemostasis, prevent bleeding, and transfuse red blood cells 1 bag at a time and only when strictly necessary); restart the antithrombotic regimen as soon as bleeding has been controlled and haemostasis has normalised; and finally, perform venous thromboprophylaxis by combining early mobilization (sitting in the preoperative period and on first postoperative day, and ambulation the day after surgery) with pharmacological measures adapted to the patient's comorbidities and chronic antithrombotic treatment. These aim of these recommendations is to improve patient comfort, hasten functional recovery and discharge home, standardise perioperative management as far as possible, and reduce the number of beds occupied by patients with OHF.

CRediT authorship contribution statement

CC contributed to the development of PICO questions and coordinated the writing of the manuscript. All authors contributed to the writing, critical review, editing, revision and final presentation of the manuscript.

All authors warrant that the questions have been thoroughly researched to obtain best practices from which all recommendations are derived. The comments and suggestions of the multidisciplinary committee of experts were also taken into consideration in the final preparation of the document.

Funding

Sociedad Española de Anestesia Reanimación y Terapia del Dolor (SEDAR).

Acknowledgements

We would especially like to thank the industrial engineer Macarena Cassinello for her help in calculating the residual effect of clopidogrel, ticagrelor, prasugrel and xabans at 24 and 48 h after the last dose; the multidisciplinary team of experts for their excellent review and contributions, and the second vice president of SEDAR, Dr. Cesar Aldecoa, and the vice presidents of the Haemostasis Division, Dr. Misericordia Basora and Dr. M.ª José Colomina for their support.