Perioperative Management of Anticoagulant Treatment in Orthopedic Surgeries

PERIOPERATIVE MANAGEMENT OF ANTICOAGULANT TREATMENT IN PATIENTS UNDERGOING ORTHOPEDIC SURGERIES

 

It is estimated that about 2% of patients receiving long-term anticoagulant treatment with vitamin K antagonists (ABK) or other anticoagulants will undergo at least once an orthopaedic procedure, emergency or elective. Almost half (~0.8%) of them will undergo hip fracture^{surgery 1}. We remind you that the main indications for which the largest number of patients receive long-term anticoagulant treatment are atrial fibrillation, the existence of a mechanical heart valve (in terms of arterial clots) and venous thromboembolic disease (in terms of venous clots).

ABCs have been for several decades the most common long-term outpatient anticoagulant treatment for the prevention and treatment of VTE and the prevention of stroke in patients with atrial fibrillation. Recently, several new oral formulations have been released with the same indications, which do not need regular laboratory testing of ABCs, have a faster action and shorter half-life, and have fewer interactions with other drugs and foods. As the number of patients taking these newer anticoagulants increases, it is evident that their perioperative management concerns more and more medical specialties and physicians.

Patients who are on anticoagulant treatment for various reasons and need to undergo surgery, in this case orthopedics, are an often intractable problem. The clinician (orthopedist, anesthesiologist, cardiologist, hematologist or other) must weigh the risk of thromboembolic and major cardiovascular events associated with the discontinuation of this treatment, against the risk of bleeding from its continuation in the perioperative period. In order to minimize perioperative complications, the associated clinician should process and answer the following questions – topics ²:

1. How to assess the patient according to the risk of thromboembolic, cardiovascular or bleeding events?
2. Is it necessary to stop antithrombotic treatment in the perioperative period?
3. If anticoagulant treatment needs to be stopped, when and for how long should it be stopped, what bridging treatment should be deemed necessary and when should anticoagulant treatment be resumed?

 

EVALUATION OF PATIENTS BASED ON THE RISK OF PERIOPERATIVE THROMBOEMBOLIC AND HEMORRHAGIC EVENTS

 

The CHADS₂ [Congestive heart failure; Hypertension; Age; Diabetes; Stroke/Transient ischemic attack] and CHA₂DS₂VASc [Congestive heart failure; Hypertension; Age; Diabetes; Stroke/Transient ischemic attack; Vascular disease; Sex category] have been validated for the assessment of the risk of embolic stroke in non-valvular atrial fibrillation in non-perioperative clinical cases, and there is no equivalent tool for perioperative assessment of the risk of embolic stroke or thromboembolism^3,4. Nevertheless, their use empirically can be extended to the perioperative environment⁴ while table 1 adapted from the ACCP of 2012 is a useful guide for the clinician⁵. In general, the indication of anticoagulant treatment, the coexisting cardiovascular and thromboembolic risk factors and the type of planned surgery should be taken into account.

class

Valve

Thromboembolic risk	Clinical indication of ABC treatment
	Atrial fibrillation	Mechanical Heart	Venous thromboembolism
High	-CHADSscore 2 5-6-Recent (<3 months) STROKE/PIE-Rheumatic valvular disease	-Any mechanical mitral valve-Older aortic mechanical valve-Recent (<3 months) STROKE/IE	-Recent (<3 months) VTHE-Severe thrombophilia1
Medium	-CHADSScore 2:3-4	-Bifold prosthetic aortic valve with at least one risk factor2	-VTE 3-12 months ago-Non-severe thrombophilia3-Recurrent VTHE-Active cancer
Mikrou	-CHADS2 score: 1-2 (without prior GNI/PIE)	-Two-fold prosthetic aortic valve without risk factor2	-FTE > 12 months

TABLE 1. Proposed risk stratification for perioperative arterial and venous thromboembolism [¹Severe thrombophilia: S, C, or antithrombin protein deficiency, antiphospholipid syndrome, or multiple abnormalities ²Risk factors: atrial fibrillation, hypertension, age > 75 years, heart failure, diabetes, stroke, or IBD ³Non-severe thrombophilia: heterozygous factor V or II mutation]

Similarly, the HAS-BLED rating [Hypertension; Abnormal renal/liver function; Stroke; Bleeding history or predisposition; Labile international normalized ratio, INR; Elderly/age >65; Drug therapy/Alcohol consumption] is used in the perioperative setting⁶ and was created and validated for the non-perioperative environment⁷. The bleeding risk in any case is largely based on the type of surgery planned. The classification of surgical procedures based on bleeding risk is undocumented knowledge in the international literature and a field of conflicting views and constant modifications⁸. Table 2, however, can in any case be used as a guide to the bleeding risk with the surgeon’s consideration. The classification of surgeries based on the bleeding risk of the Spanish Forum for Anticoagulants and Anesthesia, which is presented in Table 3^9,10, is also considered useful and guiding for the surgeon.

 

TABLE 2.

RISK OF BLEEDING DEPENDING ON SURGERY
LOW	MEDIUM	HIGH
-Dental-Dermatological-Biopsies-Minor orthopedics, ENT, general surgery, urological-Ophthalmological (anterior chamber)	-Visceral-Cardiovascular-Major orthopedics, ENT-Endoscopies	-Transurethral prostatectomy* -Intracranial neurosurgery-Intracanal-Ophthalmological (posterior chamber) -Total hip arthroplasty*

* Transurethral prostatectomy and total hip arthroplasty are not high-bleeding risk procedures for all authors

 

 

 

TABLE 3.

Low bleeding risk	-If necessary, hemostasis is feasible -A possible bleeding does not endanger the patient’s life or the outcome of the operation-Transfusion is usually not needed, e.g. minor orthopedic surgeries
Moderate bleeding risk	-If necessary, hemostasis can be difficult -A possible bleeding increases the risk of transfusion or carries the risk of reoperation E.g. major orthopedic surgeries
High bleeding risk	-A perioperative hemorrhage can endanger the patient’s life or the outcome of the operation E.g. intracranial surgeries, spinal cord surgeries, posterior chamber surgeries

 

DISCONTINUATION OF ANTICOAGULANT TREATMENT DURING THE PERIOPERATIVE PERIOD

Based on the above, and always according to the judgment of the attending physician, in minor orthopaedic procedures, such as e.g. closed realignments, percutaneous osteosyntheses, removal of superficial foreign bodies or lumps or others, the discontinuation of anticoagulant treatment is not necessary¹¹. On the contrary, in all other orthopaedic procedures, which constitute the vast majority, the modification of the anticoagulant treatment is considered necessary, because, otherwise, the outcome of the surgery and, rarely for the orthopedic specialty, the patient’s survival is endangered.

 

 

TIME TO STOP AND RESTART ANTICOAGULANT TREATMENT AND BRIDGING THERAPY

The interruption and resumption of treatment is largely determined by the nature of the surgery to which the patient is to undergo, i.e. whether it is an elective, emergency, or emergency surgery. In the first case, there is obviously the necessary time for all the necessary therapeutic interventions to stop and possibly bridge the anticoagulant therapy, the need for which will be judged by the combination of thromboembolic risk of the patient and hemorrhagic risk of the operation to which he is to be subjected as schematically depicted in Figure 1.

The necessary time probably does not exist in the second case and certainly not in the last, so there should be alternative methods of dealing with the clinical situation. In each of these clinical cases, the way the problem is managed differs depending on the anticoagulant treatment the patient was taking, i.e. whether it is ABK or one of the newer oral anticoagulants (NOACs).

 

ELECTIVE ORTHOPEDIC SURGERIES

Heparin and HTMB

Patients receiving unfractionated heparin or HPMB may undergo elective surgery without discontinuation, provided that there is a gap of 4-6 hours since the last administration for classical heparin and 12-24 hours for HBMB, depending on the prophylactic or therapeutic dose of these preparations⁵.

 

Fondaparinux (ArixtraÒ)

Patients receiving Fondaparinux can undergo elective surgery without stopping its administration, provided that 24 hours have elapsed since the last administration, and the next dose should be at least six hours after surgery, given the necessary haemostasis. Fondaparinux has no antidote while prothrombin (PT) and partial thromboplastin (APTT) times have no use in its laboratory detection.

 

Vitamin K antagonists

Warfarin has a half-life of 36-42 hours and acenocoumarol 8-11 hours. However, there is no distinction in the literature in the preoperative discontinuation time, which is for both five days before scheduled surgery to allow the elimination of its residual anticoagulant activity and the return of INR to normal, or near-normal, levels during surgery (INR< values1.5 are considered safe for surgery). This time period allows for satisfactory synthesis of anticoagulant agents II, VII, IX, and X. This regimen is effective in over 90% of patients¹², while achieving a significantly lower INR compared to a one-day preoperative discontinuation regimen in combination with vitamin^K-13 administration, and usually takes one day of discontinuation for acenocoumarol and 2.5 days for warfarin to reduce the INR from a value between 6.0 and 10.0 in 4.0³³

Prospective studies suggest that resumption of ABC therapy is possible for most patients the same evening or the morning after surgery if satisfactory hemostasis has been achieved¹⁴. Resumption of ABCs in the first 12-24 hours after surgery is not likely to lead to an increased risk of postoperative bleeding since, depending on the formulation, at least two to three days are required to achieve detectable anticoagulant effect (INR>1.3) and five to seven days to achieve therapeutic anticoagulant effect (INR>1.9)².

The need for bridging therapy, typically with EMMB, is largely determined by the patient’s estimated perioperative thromboembolic risk, as reported earlier (Table 1). The heparin bridging approach is based primarily on non-randomized observational studies² (Table 4). “High-risk” patients for perioperative thromboembolism receive “bridging” therapies with therapeutic doses of HPMB (e.g., enoxaparin 1mg/kg subcutaneously twice daily). Strategies in “moderate-risk” patients include therapeutic doses of HMBB or intermediate doses of HMMB (e.g., enoxaparin 40 mg twice daily) and no “bridging”.

group

Patient	Preoperative treatment	Postoperative treatment
Low to moderate bleeding risk	-Stopping bridging with a therapeutic dose of HMMB in the morning (20-24 hours) before surgery -Skipping an evening dose in a dosing regimen of twice a day-50% of the total daily dose in a dosing regimen once a	day-Resumption of bridging with a therapeutic dose of EMMB 24 hours after surgery and when there is sufficient hemostasis
High bleeding risk	-Stopping bridging with a therapeutic dose of HMMB in the morning (20-24 hours) before surgery -Skipping an evening dose in a dosing regimen of twice a day-50 % of the total daily dose in a dosing regimen once a	day-Resumption of bridging with a therapeutic dose of HMMB 48-72 hours after surgery and when there is sufficient hemostasis-Administration of a low dose of HMMB or avoidance of postoperative bridging

TABLE 4. Proposed perioperative treatment of bridging anticoagulant treatment

 

In general, bridging therapy is a reasonable choice in patients at high risk for thromboembolic event. In medium-risk patients, the bridging decision will be individualized according to the patient and the risk factors associated with surgery, such as the estimated risk of bleeding. Low-risk patients probably do not need bridging therapy. Characteristically, the recent Bridge⁴¹ study, which refers to patients with non-valvular atrial fibrillation, low CHADSVASC score who underwent low-bleeding risk surgeries, demonstrated that non-bridging was equally effective in preventing arterial thromboembolic events with a lower rate of bleeding. In another retrospective study of total arthroplasties⁴², bridging therapy was associated with an increased risk of bleeding-related complications.

Bridging therapy is generally initiated three days before surgery, or alternatively one day after discontinuation of asenocoumarol (Sintrom), or two days after discontinuation of warfarin (Panwarfin). If the INR remains prohibitive (>1.5) one to two days before surgery, a low dose (1 – 2 mg) of vitamin K may be administered orally¹⁵. EMMBs have a half-life of four to five hours and therefore the last dose should be given in the morning before surgery (20-24 hours). Data from observational studies suggest that when the last dose of BMB in bridging therapy is given the night before surgery, more than 50% of patients have detectable anticoagulant activity¹⁶. Therefore, if a dosing regimen is administered twice daily, only the morning dose of the day before surgery should be administered.

Postoperatively, the assessment of the bleeding risk and hemostasis of the surgical wound will dictate the resumption of bridging therapy. Evidence from observational studies suggests that resumption of fixed dosing regimens 12-24 hours after surgery may have bleeding rates of up to 20% in major surgeries¹⁷. It is generally recommended that the resumption of an HMMB bridging therapeutic dose be delayed for at least 24 hours after surgery, 48-72 hours after high bleeding risk surgeries, and 24 hours after low bleeding risk^{surgeries 2}. The clinician should decide whether to apply a therapeutic or lower-dose IMMB bridging regimen, depending on the predicted bleeding risk. Bridging therapy is discontinued once the INR rises to therapeutic levels.

 

 

Newer oral anticoagulants

For newer oral anticoagulant drugs (NOACs) there is no documentation for their perioperative management and it would be useful to know some general data:

 

1. The therapeutic dose of chronic anticoagulant treatment is significantly higher than that of thromboprophylaxis.
2. The half-life of preparations varies and increases with deterioration of renal function, which affects the time at which each preparation should be discontinued.
3. NOACs have a rapid onset of action with a maximum concentration 1-2 hours after intake
4. The residual levels of NOACs in the blood after their discontinuation that can be considered safe for surgery are currently unknown. There is no widely used or reliable laboratory test that correlates their action with bleeding risk. There is therefore no known threshold in which the bleeding risk of patients is similar to that of the general population.
5. Studies ANNEXA-A and ANNEXA-R⁴³ have recently been published according to which Andexanet is an antidote to all factor Xa inhibitors and is in the process of approval by the pharmaceutical agencies. The REVERSE-AD⁴⁴ study was also recently published, involving a monoclonal antibody, idarucizumab, which is an antidote to Dabigatran and based on this the PRAXBIND formulation was approved by the FDA and the

Given the absence of laboratory tests to accurately measure their anticoagulant activity, the perioperative administration of NOACs should be supported by:

1. In their half-life (with normal kidney function)
2. The effect of kidney function on half-life
3. The bleeding risk associated with the operation and whether the patient will undergo subarachnoid/epidural anaesthesia

Based on the above, we could distinguish the recommendations for perioperative discontinuation of NOACs into discontinuation with and without BMB bridging therapy, corresponding to ABCs (Figure 1)¹⁰.

• The Thrombosis Society of Canada recommends stopping two half-lives before surgery, i.e. one day, for all three preparations for small bleeding risk operations. In operations of high bleeding risk or in cases of subarachnoid/epidural anaesthesia, it suggests a discontinuation of 4-5 half-lives for all NOACs, which corresponds to 2-3 days for Dabigatran and two days for Rivaroxaban and Apixaban. He points out that in patients with impaired renal function, Dabigatran should be discontinued even earlier^38,39,40.
• Recent guidelines from the European Heart Rhythm Society do not recommend that patients with non-valvular atrial fibrillation be bridging with HTMB preoperatively, while they recommend that in operations with “non-clinically significant bleeding risk, the operation should be performed 18-24 hours after the last NOAC intake, for “low bleeding risk” operations last NOAC intake 24 hours preoperatively in patients with normal renal function, while in patients of “high bleeding risk” or those who are going to undergo spinal anesthesia, they suggest a 48-hour interruption. The above instructions together with those for cases of impaired renal function, especially dabigatran, are summarised in Table 5⁴⁶.

	Dabigatran		Apixaban – Rivaroxaban
	Non-clinical risk of bleeding: >12 or 24h from the last intake
	Low-risk	High risk	Low-risk	High risk
CrCl≥80mL/min	≥24h	≥48h	≥24h	≥48h
CrCl 50-80mL/min	≥36h	≥72h	≥24h	≥48h
CrCl 30-50mL/min	≥48h	≥96h	≥24h	≥48h
CrCl 15-30mL/min	Contraindicated	Contraindicated	≥36h	≥48h
CrCl < 15mL/min	Contraindicated

Table 5.

• Other guidelines^9,10,19, suggest as a safer approach in elective surgeries, bridging therapy with EMMB during discontinuation of NOACs, especially in patients and procedures of high thromboembolic risk. With this approach, anticoagulants are discontinued five days before any operation, i.e. more than five halves of life of all preparations. This prolonged interruption offers enough time to eliminate plasma NOACs levels even in elderly patients or patients with renal impairment. This approach is recommended for patients at moderate to severe thromboembolic and haemorrhagic risk and is based on empirical data rather than evidence-based knowledge.

Bridging with BMB is done in these patients in a similar way to bridging for ABK (Figure 2)¹⁰. Initiation of EMMB may occur 12 hours after the last dose of NOAC at twice daily dose, or 24 hours after, at once daily^{administration 19}. The last dose of IMMB will be 24 hours before surgery in a therapeutic dosing regimen, or 12-24 hours in a prophylactic dosing regimen.

 

Bridging therapy Moderate thrombotic or bleeding risk CrCl >50 ml/min
LAST DOSE OF NOAC	FIRST DOSE OF HMB	HMMB	HMMB	LAST INSTALMENT HMMB*	C/O	NOAC OR HMB (WHEN HAEMOSTASIS IS ACHIEVED)
-5	-4	-3	-2	-1	0	+1	+2
			LAST DOSE OF NOAC	NO NOAC	C/O	FIRST POSTPARTUM DOSE NOAC	NOAC AT THE APPROPRIATE DOSE
No bridging treatment Low bleeding or thrombotic risk. CrCl >50 ml/min

FIGURE 2. Scheme of recommendations based on and modified by the Spanish Forum on Anticoagulants and Anesthesia.

*The last anticoagulant dose of EMMB will be 24 hours before surgery (half dose if given daily) or 12-24 hours if given a dose of thromboprophylaxis

 

The ideal timing for resumption of NOAC will depend primarily on postoperative bleeding risk, with the first dose being administered once hemostasis is achieved. Most recommendations agree on a start of 24-48 hours postoperatively, again depending on the magnitude of the bleeding risk of the operation the patient underwent (24 hours for small operations and 48-72 hours for high bleeding risk operations)^{38, 39, 40} possibly even at half dose^10.40. Alternatively, HMMB may be given postoperatively with resumption of NOAC at full dose on the third or fourth postoperative day and 24 hours after HMMB¹⁰. It should be stressed that under no circumstances should there be a co-administration of HTMB and NOAC, unlike the practice in coumarin, due to the rapid action of the last¹⁹.

 

EMERGENCY AND EMERGENCY ORTHOPEDIC SURGERIES

Classical heparin

Patients receiving classical heparin can undergo surgery without interruption of administration, provided that 6-12 hours elapse from the last subcutaneous administration. In an emergency or life-threatening bleeding, the antidote to classical heparin is protamine sulfate, a protein that has a mild anticoagulant effect on its own. However, when heparin is administered, it binds to it, inhibiting its anticoagulant^{effect 26}. The inhibitory effect of protamine sulfate is very fast (within a few minutes) as its half-life is seven minutes. One mg of protamine sulfate neutralizes about 100 units of heparin within 15 minutes. The dose of protamine sulfate in cases of continuous intravenous administration of heparin is calculated according to the dose administered during the last 2-2.5 hours. In contrast, the administration of protamine sulfate in subcutaneous administration of heparin is of a longer duration due to the longer duration of action of heparin. The action of the antidote can be assessed in the laboratory by measuring aPTT²⁶.

 

Small Molecular Weight Heparins (HPMB):

Patients receiving HPMB for prophylaxis may undergo surgery without discontinuation, provided that 6-12 hours elapse from the last dosing depending on the formulation. In case they receive a therapeutic dose, 12-24 hours should intervene. HPMBs do not have a specific antidote. Protamine sulfate is not their antidote as it reverses their action only partially, neutralizing, as shown in studies, more anti-IIa activity than anti-Xa action^27-31 . The first measure in reversing the action of EMMBs is to stop issuing them. In clinical cases of emergencies or bleeding, where the anticoagulant effect of EMMBs must be reversed, protamine sulphate is administered at a dose of 1 mg for 100 units of anti-Xa up to a maximum dose of 50 mg (e.g. 1 mg of enoxaparin corresponds to 100 units of anti-Xa) as long as the administration of EMMB has taken place within eight hours. If bleeding continues, a second dose of 0.5 mg of protamine sulphate per 100 units of anti-Xa is administered. If the HPMB has been administered in more than 8 hours, smaller doses of protamine sulphate may be administered. Under these conditions, even with large doses of protamine, anti-Xa activity is never completely neutralized (up to 60% in the case of enoxaparin) and some antithrombotic activity is maintained. Laboratory measurement of anti-Xa levels allows the evaluation of the activity of HPMB but cannot be used to evaluate the anti-reversal effect of protamine sulphate on HPMB ³².

 

Vitamin K antagonists

In cases of emergency operations, where there may be a wait of 24-48 hours, discontinuation of ABCs in combination with the administration of their antidote, vitamin K, is able to normalize the INR in the desired time period (24-48 hours)²¹. Vitamin K reverses the action of ABCs as it competes with their toxic activity with the formation of coagulation factors II, VII, IX, and X. The fastest way to reduce INR is to administer a large amount of vitamin K intravenously³⁴. A small amount of vitamin K intravenously has the same effect in 24 hours as oral administration of Vitamin K³⁵. The dosage regimen of vitamin K is 1-10mg intravenously in a period of 20-30 minutes. The same dosage can also be administered orally.

If there is ongoing, life-threatening bleeding, or if the goal is to restore coagulation in order to undergo an emergency operation, it is necessary to co-administer with vitamin K (5-10 mg slowly intravenously) and two to four units of Fresh Frozen Plasma (FFP) or active Prothrombin Complex Concentrate (FFP) aPCC)^21,47. If there is an urgent or life-threatening case, guidelines from the ACCP suggest administering aPCC and intravenous vitamin K, 5-10 mg, instead of FFP and vitamin K²⁰. The advantages of aPCCs over FFP include a smaller injectable volume, faster administration, and the administration of a more reliable amount of vitamin K-dependent clotting agents, while their disadvantage is the increased cost. These properties make concentrated prothrombin complexes (and mainly activated 4-factor complexes containing factors II, VII, IX and X, as well as proteins C and S) the agents of choice for the reversal of coumarin anticoagulants in patients who require emergency surgery, or are bleeding actively and need coagulation factor substitution²¹. The usual average dose is 30 units/kg body weight (maximum 3000 units)²² but it is modified according to INR. It should be emphasized that the antithrombotic effect of anticoagulants is the result of the immediate inhibition of the action of coagulation agents and not of their reduced quantity. Therefore, the efficacy of non-specific haemostatic agents is limited, their use is empirical and not documented in the literature by clinical studies.

 

Fondaparinux

In case of emergency or emergency surgery, a 24-hour delay would be the best option. If this is not possible, discontinuation of treatment is the first step, while the use of recombinant factor VIIa (rFVIIa) is an option to control possible bleeding⁴⁵.

 

Newer oral anticoagulants

As mentioned above, there is no widely used or reliable laboratory test that correlates their action with bleeding risk, the half-life of preparations vary and increase with deterioration of renal function, while all preparations have a rapid action with a maximum concentration of 1-2 hours after intake.

As regards the calculation of the anticoagulant activity of NOACs, standard assays can be used to make a qualitative assessment of their presence. Thus, a normal thrombin time (TT) excludes the presence of Dabigatran, and a normal partial thromboplastin time (APTT) makes it unlikely to be involved in bleeding⁴⁰. Similarly, a normal prothrombin time (PT) value (with thromboplastin sensitive to rivaroxaban) indicates not high levels of the drug, but does not exclude its presence. Rivaroxaban does not prolong TT⁴⁰, and there are no literature data for apixaban.

In general, it is recommended to delay surgery, if possible, until the coagulation tests mentioned are normal or sufficient time has elapsed for the elimination of each preparation⁴⁰. If this is not possible, some literature data suggest the use of aPCC, or concentrated factor X (for anti-Xa preparations) but as there are no clinical studies to support this, their use is recommended to be limited to intracranial or life-threatening hemorrhages⁴⁰. The short half-life of these drugs makes the need to administer PCCs or other prohemostatic agents unnecessary in most clinical cases^23,24.

It is also very important to know the exact time of the last dose, in order to assess the perioperative risk and determine the appropriate time for surgery, while if the last administration is within the previous two hours, the use of oral activated charcoal^39.40 is recommended. Especially for dabigatran alone, dialysis can be applied in case of prolonged anticoagulant action, relying on the relatively low protein binding of^18.25.

It is worth noting that in the second half of 2015, phase III studies were published for two antidotes, Dabigatran and all anti-Xa agents, Idarucizumab and Andexanet Alfa respectively. The former has already received approval for use by the FDA and EMA (Praxibind®), and the latter is in the process of being approved. In these studies, the antidotes demonstrated a very rapid, within a few minutes, reversal of the action of anticoagulant drugs and it is reasonable to assume that the aforementioned methods of perioperative management in cases of emergency and emergency operations will not have clinical application.