Management of Antithrombotic Drugs in Association with Endoscopic

Home , Argatroban, Ditazole, Iloprost, Indobufen, Trapidil, Triflusal

SPECIAL ARTICLE

Management of antithrombotic drugs in association with endoscopic procedures* Fernando Alberca-de-las-Parras1, Francisco Marín2, Vanessa Roldán-Schilling3 and Fernando Carballo-Álvarez1 1Unit of Digestive Diseases and 2Department of Cardiology. Hospital Clínico Universitario Virgen de la Arrixaca. IMIB. Murcia, Spain. 3Department of Hematology and Oncology. Hospital Universitario Morales Meseguer. Murcia, Spain

ABSTRACT benefits in terms of both financial savings and thrombosis incidence (9). The use of antithrombotic drugs (anticoagulants and antiplatelets) A first Spanish version of these guidelines was pub- has increased significantly with our understanding of cardiovascular risk. Encountering patients on these therapies who require an lished in 2010 (6). However, the emergence of new drugs endoscopic procedure is therefore increasingly common. At and of significant data in the literature has prompted a decision making the endoscopist must rely on other specialists revised and updated version, for which the consensus of (basically cardiologists and hematologists) as risk not only lies among the Working Group of Thrombosis of Spanish Cardiology increased bleeding odds but also in the possibility of thrombosis Society (GTCV-SEC) and the Spanish Society of Throm- following dose discontinuation or change. Understanding the pharmacology, indications, and risks of endoscopic procedures is bosis and Hemostasis. therefore essential if sound decisions are to be made. The efforts An analysis of the scarce adherence to guidelines and of four scientific societies have been brought together to provide of their applicability (10,11) suggests that less academic, clinical answers on the use of antiplatelets and anticoagulants, as more practical guidelines should be developed, since the well as action algorithms and a practical protocol proposal for endoscopy units. criteria for early drug re-introduction are approached by cardiologists (12), which conditions follow-up by gastro- Key words: Antithrombotic drugs. Antiplatelets. Anticoagulants. enterologists for fear of early bleeding. Endoscopic procedures. The goal of the present paper is to summarize the available evidence, and to discuss newly marketed medications using a clinical questions and answers approach to help INTRODUCTION gastroenterologists in their decision-making.

For many years single or dual anticoagulation and antiaggregation have become a therapeutic and preventive WHAT MANAGEMENT WOULD BE tool, particularly regarding highly prevalent cardiovascu- APPROPRIATE FOR ACUTE GI BLEEDING lar diseases. This, together with increased interventionism IN PATIENTS ON ANTICOAGULANTS OR among endoscopic techniques, represents a higher risk. ANTIPLATELETS? However, such risk is not only derived from potential bleeding but also from thrombotic risk (1), its obviation As is usual in medical practice, decision making de- by endoscopists being unacceptable. pends upon risk and severity. In patients with high throm- Scientific societies have issued a number of guidelines botic risk (for example, with mitral prosthesis) and mild and recommendations for such situations (2-8) and com- bleeding anticoagulation withdrawal is inappropriate pliance with them have even brought about secondary –pharmacological and endoscopic measures may be considered while leaving reversal aside. *Guidelines endorsed by the Working Group of Thrombosis of Spanish Car- In general, anticoagulants must be discontinued imme- diology Society (GTCV-SEC), Spanish Society of Gastrointestinal Endos- diately for severe bleeding. In the absence of severity cri- copy (SEED), Spanish Society of Digestive Diseases (SEPD) and Spanish Society of Thrombosis and Hemostasis. teria the approach should hinge upon the patient’s throm-

Received: 25-01-2015 Accepted: 16-02-2015

Correspondence: Fernando Alberca de las Parras. Unit of Digestive Dis- Alberca-de-las-Parras F, Marín F, Roldán-Schilling V, Carballo-Álvarez F. Man- eases. Hospital Clínico Universitario Virgen de la Arrixaca. Ctra. Madrid- agement of antithrombotic drugs in association with endoscopic procedures. Cartagena, s/n. 30120 El Palmar. Murcia, Spain Rev Esp Enferm Dig 2015;107:289-306. e-mail: [email protected] 290 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

botic risk, and the following therapies may be initiated, 4. Recombinant activated factor VII (Novoseven®): whether sequentially or in parallel with the resolutive en- 90 µg/kg in slow bolus (2-ml ampoule = 1.2 mg). doscopic technique (3,13) (Fig. 1). Effects are seen at 10 to 30 minutes after dosing, 1. Vitamin K1, IV (phytomenadione), 10 mg (one am- and last 12 hours. It normalizes prothrombin time poule) in 100 ml of saline or 5% glucose solution: 10 ml and fixes platelet malfunction. It is highly thrombo- over 10 minutes (1 mg/10 min), the rest in 30 minutes. genic, and its use is currently indicated in hemophil- The effect begins within 6 hours and lasts for 12 hours. iacs with inhibitor, acquired hemophilia, and severe 2. Fresh plasma, 10 to 30 ml/kg; it may be repeated in congenital FVII deficiencies only. The latter two halved dose after 6 hours, as factors have a half-life drugs may not be administered together. of 5-8 hours. Currently, guidelines do not recom- These measures will be applied until the INR becomes mend the use of fresh plasma to reverse anticoagu- stable between 1.5 and 2.5, and bleeding stops. lation except when it is the only measure available, The bleeding risk of patients on anticoagulants is 6.5- nor is this therapy accepted for plasma expansion. fold higher than in those off these drugs (14). 3. Prothrombin complex concentrate containing fac- From a practical standpoint, and in the emergency set- tor IX ((BERIPLEX 500 IU®, OPTAPLEX®, PRO- ting, the right thing to do is to initially use a prothrom- THROMPLEX IMMUNO TIM 4 600 IU®): Equiv- bin complex concentrate, which is the fastest option to alent to 500 ml of plasma. Dosage is estimated achieve reversal. using the following formula: (Desired prothrombin The approach modestly changes with the new oral time - obtained prothrombin time) x kg weight x 0.6. anticoagulants (NOACs) as neither vitamin K nor fresh It may be thrombogenic. plasma have proven effective (15). Attitude will logically

ACUTE GI BLEEDING IN AN ANTICOAGULATED PATIENT

NOAC ACENOCOUMAROL/WARFARIN

Mild Moderate Severe Mild bleeding Severe bleeding bleeding bleeding bleeding

Thrombotic risk WITHDRAWAL + +

Withdrawal Specific management Prothrombin (endoscopy) complex High Low Sequential therapy Platelets if < 60,000 if concentrate dabigatran, keep adequate urine Vitamin K output No withdrawal Prothrombin complex concentrate

Achieve reversal of INR to 1.5-2.5

No risk of relapse

Reinitiate anticoagulants (1-5 days) +/- use of IV heparin sodium if high thrombotic risk

Fig. 1. Algorithm for acute bleeding in patients on anticoagulation.

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 291

depend on bleeding severity, paying particular attention erally speaking, only severe FVII and FX deficiencies to renal function especially in patients on dabigatran, as need correction (< 10%). it prolongs its anticoagulating action. This is also the case Activated partial prothrombin time (APTT) is mea- with apixaban and rivaroxaban in hepatocellular dysfunc- sured as an extension of PT. It may be seen in the (de- tion. creasing) use of heparin sodium i.v. and in some patients Antiplatelets and NSAID discontinuation will not al- with LMWH. As regards heparin sodium APTT becomes ter the natural history as effects on hemostasis will linger normal after discontinuation within 4 to 6 hours. How- for a few days; however, we should bear this in mind for ever, should faster reversal be desirable in an emergency lesions that predictably will not heal fast. The fact that as- setting, protamine sulphate is to be used. As for LMWH, pirin and NSAIDs may also be a cause of bleeding is not the effect usually clears within 12-24 h, and APTT nor- to be forgotten. Their effect may, nevertheless, be partly malization may take slightly longer. In the case of LMWH reversed with fresh platelet transfusions when advisable protamine is of limited help. APTT prolongation may also according to bleeding severity. be seen in factor deficiencies (hemophilia A/B/C), and in the presence of lupus anticoagulants.

WHAT COAGULATION CHANGES SHOULD BE CONSIDERED BEFORE ENDOSCOPY AND HOW WHAT ENDOSCOPIC TECHNIQUES ARE CAN THEY BE CORRECTED? (TABLE I) CONSIDERED TO BE ASSOCIATED WITH A HIGHER BLEEDING RISK? In extreme situations (fewer than 10,000 platelets, INR higher than 6, etc.) with active bleeding, we must cor- Bleeding risk has been stratified by expert consensus, rect such changes before the endoscopic technique, as this as indicated in clinical guidelines (2) and manuals, into may stop the bleeding and render the technique –and its two risk levels (Table II). Basically agreeing with this potential complications– unnecessary. classification, we may importantly increase risk strat- In patients with thrombocytopenia our attitude will de- ification, as an endoscopic sphincterotomy in a patients pend on procedure type: for high-risk procedures platelet with multiple conditions is clearly unlike the excision of count must be higher than 50,000/μl; for low-risk proce- a pedicled polyp. It is therefore important that the endos- dures a platelet count > 20,000/μl is acceptable. Plate- copist’s view be taken into account particularly after the let deficiency is corrected on a schedule using infusions procedure when considering antiplatelets/anticoagulant during the technique and immediately before it. It suffices reintroduction times. with increasing platelet numbers to 40,000-50,000/mm3. – Colonoscopy: This is a technique that in itself does Each platelet unit increases the count by 5,000-10,000 per not increase bleeding risk in patients with coagula- mm3, and the dose is 1 U/10 kg of body weight. In case of tion issues. platelet refractoriness with severe bleeding evidence the – Biopsy: Bleeding risk during biopsy taking is ex- use of recombinant activated factor VIl (rVlla) (Novosev- tremely rare, approximately 1‰ (16). en®) is recommended, with hemostatic action in patients – Polypectomy: Polypectomy-related major bleeding with severe thrombocytopenia and thrombocyte disease, risk in several extensive series oscillates between at a dose of 90 to 150 µg/kg every two hours intravenous- 0.05% and 1%, but it may be up to 4.3% for polyps ly until bleeding resolution. greater than 1 cm, and 6.7% for those larger than Prothrombine activity lower than 50% (vs. INR > 1.4) 2 cm (17-19). This risk seemingly increases for pol- is usually found in chronic liver disease, factor VII and yp sizes above 2 cm, and pedunculated polyps have X deficiencies, and oral anticoagulation. In such case the been seen to bleed more severely than sessile ones, above schemes are used for anticoagulated patients. Gen- hence associating a risk-reducing technique such as

Table I. Drug-related hemostatic changes Agent Normal hemostasis Prot. time/INR PTTA Antiplatelets 2-10 days Normal Normal NSAIDs 1-3 days Normal Normal IV heparin sodium 2-4 hours Longer * Longer +++ LMW heparin 12 hours Normal Normal Oral anticoagulants 4-6 days Longer +++ Longer * Normal Normal NOACs 12 h-4 days Longer + with rivaroxaban Longer + with rivaroxaban

Rev Esp Enferm Dig 2015; 107 (5): 289-306 292 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

Table II. Bleeding risk during endoscopic procedures Low (< 1%) High (1-6%) Diagnostic endoscopy with or without biopsy Polypectomy Diagnostic endoscopic retrogade cholangiopancreatography Laser coagulation and ablation Biliary stent insertion without sphincterotomy Sphincterotomy Echoendoscopy Benign or malignant stenosis dilation Push enteroscopy Percutaneous endoscopic gastrostomy EUS-FNAP Balloon enteroscopy Submucosal dissection and mucosectomy

the use of prophylactic adrenaline, endoloop, and In an extensive registry of 120,886 procedures in 95,807 endoclips is advisable for patients with coagulation patients bleeding was only identified in 2% of patients issues (19). having discontinued heparin (27). Another paper only Two types of bleeding exist: Immediate bleeding describes 3 bleeding episodes among 225 polypectomies when nurturing vessels are improperly coagulated, and for 123 anticoagulated patients where anticoagulants had delayed bleeding, after 1 to 14 days, which is more se- been discontinued previously (28). Recently, a paper was rious; this should be considered before antiplatelets re- reported where polyps smaller than 10 mm were excised introduction. in 70 patients on acenocoumarol, which found higher In antiaggregated patients its role for GI bleeding has bleeding rates in the group treated with a hot loop versus been poorly analyzed; in an animal study, 7-mm lesions a cold loop, both regarding early (23% vs. 5.7%) and de- in the colon showed an increase in bleeding time among layed (14% vs. 0%) bleeding (29). those on aspirin vs. no therapy (155 seconds vs. 169 ERCP and sphincterotomy. The incidence of bleed- seconds – p < 0.05) (20); similarly bleeding has been ing falls between 2.5% and 5% (30). Different studies found to be increased at biopsy sites in patients on aspi- define anticoagulation as a clear risk factor for bleed- rin versus controls or NSAIDs (21). However, these ex- ing. Regarding antiaggregation only two retrospective perimental studies have found no clinical application in studies (31,32) have discussed this effect, with incon- the literature regarding patients undergoing endoscopic clusive results as considered in the American Society’s procedures. clinical guidelines, but with a risk for acute bleeding in Aspirin and NSAIDs have not been shown to increase one of them (32) of 9.7% vs. 3.9% (ASA vs. controls, p < bleeding risk during polypectomies, albeit these are the 0.001), and a risk for delayed bleeding of 6.5% vs. 2.7% retrospective or limited-analysis studies on which both (p = 0.04), which suggests that, except for emergency the ASGE and Euroean guidelines were based; these settings, discontinuation is desirable at least 7 days be- guidelines do not recommend antiplatelet discontinu- fore the technique. Balloon sphincteroplasty has been ation before endoscopy (22). However, a prospective proposed as an alternative technique for patients in need study amongst them, while not showing increased com- of urgent bile duct opening and with coagulation disor- plications, does find trace amounts of blood in the stools ders (33), as well as in the transient use of biliary pros- of up to 6.3% of patients versus 2.1% of subjects on theses without sphincterotomy. placebo (23). The most relevant complication, name- Enteroscopy. The simple technique seems to pose no ly delayed bleeding, was specifically discussed in a risk whatsoever, but should be considered a risky tech- case-control study, thus confirming no increased bleed- nique when used as a therapeutic, interventional approach, ing in the subgroup on aspirin versus placebo; however, which is the case in up to 64% of cases (34). the use of aspirin was not studied beyond 3 days fol- Endoscopic submucosal dissection (ESD). In a recently lowing the procedure, and bleeding episodes have been reported retrospective paper in patients undergoing 1,166 described up to 19 days afterwards (24). Hence, some ESDs delayed bleeding was analyzed, and the use of anti- authors recommend that antiplatelets be discontinued thrombotic medications was shown to be an independent within 4 to 7 days of risky techniques, and reintroduced risk factor, higher still when two drugs were used (no sta- after 7 days when used as secondary prevention, or 14 tistical significance) (36). days when used as primary prevention, following pol- Percutaneous endoscopic gastrostomy (PEG). Sev- ypectomy (25). eral papers exist that retrospectively discuss the role of In anticoagulated patients bleeding risk after polypec- antithrombotics in this procedure, and fail to detect any tomy is increased, even when anticoagulation is discontin- increased bleeding risk when aspirin and clopidogrel are ued before the procedure (2.6% vs. 0.2%, p = 0.005) (26). used, even concomitantly (36).

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 293

WHICH ARE THE PRIMARY ANTIPLATELETS – Unstable angina or AMI without ST elevation not AVAILABLE IN SPAIN? requiring coronariography: DA for 1 to 12 months. – Metallic stent placement: DA up to 12 months. Antiplatelets are drugs that act by inhibiting platelet – Antiproliferative drug-sluting stents: DA for 12 functioning at various levels, as seen in figure 2. Its classi- months. fication depends on their mechanism of action (Table III). – After fibrinolysis (up to 12 months). There is a wide spectrum of actions, and both antiplate- – Stable coronary disease: Only aspirin (clopidrogrel lets effects and duration seem to vary for each one of them in aspirin intolerants). (Table IV). – Heart disease primary prevention in patients at risk: A factor for consideration is their association with non- No DA necessary, only aspirin or clopidogrel, which steroidal anti-inflammatory drugs (NSAIDs), which may may be discontinued without risk. enhance antiaggregating effects. – AMI without reperfusion therapy (at least 1 month, The novel ADP receptor blockers are key to the prevention ideally up to 12 months). of acute coronary syndrome, mainly following the placement – Ischemic stroke associated with large-vessel and of stents. From a gastroenterologist’s standpoint, we should small-vessel arteriosclerosis (aspirin, cilostazol or know their antiaggregation effect is immediate and stronger clopidogrel) (39). than that of aspirin, and therefore we should discontinue them  within 5 to 7 days of a high-risk procedure (37) (Table V). WHAT IS THE RISK OF DISCONTINUING A PROPERLY INDICATED ANTIPLATELET WHAT ARE THE PRIMARY INDICATIONS THERAPY? OF ANTIPLATELETS? Early dual (aspirin + clopidogrel) antiplatelet therapy As gastroenterologists we must now these indications, discontinuation entails a stent thrombosis risk of 29% as well as the impact that the discontinuation of a properly (8%-30%) (40). Discontinuing dual antiaggregation after indicated drug may have for patients (8). stent implantation increases the risk for potentially-le- Indications include (38): thal subacute stent thrombosis by 15% to 45% within a – AMI with or without ST elevation: dual antiaggrega- month. In the secondary prophylaxis of AMI risk is higher tion (DA), aspirin plus clopidogrel (or prasugrel or (3-fold) for discontinuation within 14 days. Risk also in- ticagrelor), for 12 months. creases within 3 days of balloon angioplasty (41).

PAR-1 Thrombin PAR-4 Clopidogrel Ticlopidine P2Y1 P2y12 Prot. G Pepducins ADP cAMP

GPIIb-I IIa Fgno. Dipyridamole Phosphodiesterase Cilostazol Dense granule AMP

Arachidonic acid Eptifatide PLATELET Tirofiban Aspirin Cyclooxygenase Abciximab Trifusal

Thromboxane A2 TXA2R

Fig. 2. The mechanism of action of antiplatelets.

Rev Esp Enferm Dig 2015; 107 (5): 289-306 294 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

Table III. Classification of antiplatelets Types Drugs Platelet adhesion inhibitors Von Willebrand factor inhibitors Acetylsalicylic acid (AAS®, Adiro®, Bioplak®, Therasacard®, Tromalyt©, generic) TXA2 production inhibitors Indobufen Triflusal (Anpeval®, Disgren®, generic) Sulfinpyrazone Dipyridamole (Persantin®) Ditazole Act on platelet activation mediators Phosphodiesterase inhibitors Pentoxifylline Cilostazol (for intermittent claudication) Trapidil Ticlopidine (Ticlid®) Clopidogrel [Plavix®, Iscover®, Duoplavix®, (Clopidogrel 75 + ADP receptor antagonists aspirin 100)] (thienopyridines) Ticagrelor (Brilique®) Prasugrel (Effient®) Abciximab Intravenous use Eptifibatide Tirofiban (Aggrastat®) GP IIb/IIIa receptor antagonists Xemilofiban Oral use Orbofiban Lotrafiban Epoprostenol (for pulmonary hypertension) Analogs of natural platelet aggregation inhibitors Iloprost (for peripheral artery disease)

Table IV. Duration of antiaggregation and activity ARE ALL ANTIPLATELETS ALIKE IN of antiplatelets and NSAIDs ANTIPLETELET POWER AND SECONDARY BLEEDING RISK? DO ALL ANTIPLATELETS Antiplatelets NSAIDs DOSES BEHAVE THE SAME WAY? 10 d. Ticlopidine Piroxicam, tenoxicam, Prasugrel Strong (7 d.): indomethacin, ketorolac, Aspirin does not contraindicate any diagnostic or ther- 7 d. Triflusal flurbiprofen apeutic procedure (7). Clopidogrel No differences have been found between aspirin 5 d. Aspirin Meloxicam, sulindac, 100 mg and 300 mg in antiplatelet power; only aspirin Ticagrelor Weak (12 h.): nabumetone, diflunisal, 300 mg in three daily 100 mg doses has a somewhat higher antiplatelet strength, hence no differential management Dipyridamole paracetamol, dipyrone 1 d. Tirofiban is warranted for patients according to their dosage (42).

Table V. Pharmacodynamics of new antiplatelets Clopidogrel Prasugrel Ticagrelor Class Thienopyridine Thienopyridine Triazolopyrimidine Reversibility Irreversible Irreversible Reversible Activation Prodrug. Limited by metabolism Prodrug. Not limited by metabolism Active drug Onset of effect 2-4 h 30 minutes 30 minutes Duration of effect 3-10 days 5-10 days 3-4 days Withdrawal before major surgery 5 days 7 days 5 days

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 295

No prospective, randomized studies in the non-cardiac actions to be taken with antiplatelets. The cut-off is set in surgery setting have been carried out to assess the risk for 14 days to avoid elective procedures or to undertake them surgical bleeding in association with clopidogrel alone or with aspirin alone; waiting time should be 30 to 45 days plus aspirin in stent carriers, and increased bleeding has for angioplasty with metallic stents, and 1 year for cov- only been described for associated aspirin doses in excess ered stents (15). of 200 mg. The indication of a bridge therapy with LMWH in an- Ticagrelor and prasugrel seemingly have a greater antiaggregated patients undergoing endoscopy with an add- tiplatelet effect as compared even with clopidogrel, hence on procedure is now much debated. No evidence exists the bleeding risk related to these new antiplatelets would for its benefits (40), and the fact that it will not effectively be higher. protect from thrombosis in patients with coronary stents In view of this information -with a low evidence level- seems clear, although its potential protective role in situ- clopidogrel and ticagrelor should be discontinued 5 days, ations at risk of thrombosis and bleeding with relatively and prasugrel 7 days, before any procedure with high undelayable endoscopies has been suggested because of bleeding risk (43). its antiaggregating role. Upon antiplatelet drug reintroduction we should re- assess both bleeding and thrombosis risks. Initially, they IS ANTIAGGREGATION, WHETHER SINGLE will be reintroduced at 24 hours after the endoscopic pro- OR DUAL, A CONTRAINDICATION FOR ANY cedure, except in situations with high bleeding risk, where ENDOSCOPIC TECHNIQUE? a cardiologist or neurologist or hematologist should be in- cluded in the assessment. No convincing data are available that turn single an- Based on this we developed a working paper to aid cli- tiaggregation into an absolute contraindication for any nicians in their decision-making before endoscopy pro- endoscopic technique (7), but ASA –where this algorithm cedures, delineating a procedure’s bleeding risk and then may be used somewhat safely– and novel antiplatelets defining its thrombotic risk (Fig. 3). –which may require a more careful assessment– should probably be considered on an individual basis. As regards dual antiaggregation no papers discuss its WHAT IS THE ROLE OF A GLYCOPROTEIN actions, although in view of the current data intervention- IIb-IIIa INHIBITOR IN THE REVERSAL OF al procedures should likely be avoided during its use. At ANTIAGGREGATION? any rate, the decision should be made based on endoscopic and thrombotic risk: In the presence of low endoscopic In a prospective, non-controlled study of 30 patients risk both drugs may be used; regarding high risk, this will undergoing undelayable major surgery within 6 months hinge upon thrombotic risk –if low, clopidogrel will be following the placement of a coated stent a glycoprotein discontinued and the procedure will be performed under IIb-IIIa inhibitor, tirofiban i.v. (Fig. 4), was used as bridge aspirin; when high, attempts will be made to delay the therapy, with clopidogrel being discontinued 5 days be- endoscopic procedure or a bridge therapy may be consid- fore, with no resulting coronary or bleeding condition: ered with glycoprotein IIb-IIIa inhibitors. this represents an interesting possibility for this subgroup of patients (44). However, no other papers have been reported on this topic, hence we must follow on cautiously WHAT IS IN PRACTICE THE REGIMEN USED and consider it an option only in the setting of joint pro- TO REVERSE ANTIAGGREGATION? tocols involving cardiology and hematology departments.

In all cases delaying therapy until thrombotic risk is minimal is most desirable; in cases on dual antiaggrega- WHICH ARE THE PRIMARY tion, clopidogrel can be discontinued 7 days before while ANTICOAGULANTS AVAILABLE IN SPAIN maintaining aspirin as per the prescribed regimen. (OTHER THAN HEPARINS)? (TABLE VI) The different guidelines (6) have posited suggestions for managing antiaggregation that are mainly based on defin- Figure 5 shows where primary anticoagulants act. Most ing whether indication is urgent, and assessing extant indi- common options, warfarin and acenocoumarol, are well cations: For primary and secondary indications, consider- known and possess a longer half-life, which may lead to ing their low thrombotic risk, aspirin may be withdrawn; consider bridge therapies with low molecular weight hep- aspirin is to be maintained for all other scenarios. arin in settings with bleeding risk. In a consensus paper between the American College of New oral anticoagulant (NOAC) drugs have been mar- Cardiology and the American College of Gastroenterolo- keted of late, including direct thrombin inhibitors (dab- gy time periods are established from the time of coronar- igatran) and factor Xa inhibitors (rivaroxaban and apix- iography to stent implantation or angioplasty regarding aban). Edoxaban remains unavailable as of today. These

Rev Esp Enferm Dig 2015; 107 (5): 289-306 296 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

ENDOSCOPY IN ANTIAGGREGATED PATIENTS

Endoscopy with high bleeding risk Endoscopy with low bleeding risk

Thrombosis risk Perform endoscopy

LOW: MID: HIGH: Primary prevention uncovered stents (1-3 m) uncovered stents (1 m) Uncovered stents (> 3 m) covered stents (12-15 m) covered stents (12 m) Covered stents (> 15 m) acute coronary syndrome +/- stent (12 m)

Aspirin or Clopidogrel Clopidogrel or ticlopidine Dual antiaggregation trifusal

Perform Suspend Tx Suspend Tx Delay Delay Assess bridge or or endoscopy +/- replace +/- replace endoscopy endoscopy therapy with +/- Tx with aspirin with aspirin tirofiban suspension

Perform Perform Perform endoscopy endoscopy endoscopy with aspirin

Fig. 3. Antiplatelet discontinuation schedule.

Table VI. Classification of anticoagulants (except heparin) -5 -4 -3 -2 -1 -0 +1 +2 +3 +4 +5 Type Drugs Acenocumarol (Sintrom®) Procedure Vitamin K inhibition Warfarin (Aldocumar®) Tecarfarin CLOPIDOGREL Dabigatran (Pradaxa®) Initiate with oral diet Desirudin 300 mg/24 hrs, then 75 mg/day Direct thrombin inhibitors Lepirudin Argatroban ASPIRIN Bivalirudin Rivaroxaban (Xarelto®) Direct factor Xa inhibitors Apixaban (Eliquis®) Edoxaban (Savaysa®) Selective factor Xa inhibitors Fondaparinux (Arixtra®) TIROFIBAN 0.4 mcg/kg/min/30 End 6 hrs after min, then 0.1 mcg/ (*) onset of clopidogrel kg/min. (*) drugs have no specific antidote, therefore it is advisable -4h +2h that sites implement action protocols for the management If renal failure (CrCl < 30 mL/min), reduce dosage by of bleeding complications, surgery preparation, and inva- 50% and double both discontinuation time (8 hrs) and sive procedures in patients receiving NOACs. Their anti- reintroduction time (4 hrs) with respect to the procedure coagulant effect lasts for approximately 24 hours, but may No treatment Treatment be longer in the presence of renal failure (up to 4 days), each one having a different clearance rate. Knowing this Fig. 4. Tirofiban bridge therapy schedule. is therefore crucial to establish a NOAC discontinuation

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 297

TISSUE FACTOR ENOXABAN RIBAROXABAN FONDAPARINUX APIXABAN PLASMA CLEARANCE CASCADE EXTRINSIC PATHWAY (factor VII, tissue factor) PROTHROMBIN (factor II) Coagulation factor activation in the liver FACTOR Xa (mediated by vitamin K) THROMBIN (factor IIa)

INTRINSIC PATHWAY (factors IX, XI, XII) LMWH DABIGATRAN

ACENOCOUMAROL BIVALIRUDINA WARFARIN FIBRIN FIBRINOGEN

THROMBUS

Fig. 5. The mechanism of action of anticoagulants scheme before an interventional procedure. While they Dabigatran and rivaroxaban increase the risk of GI prolong bleeding time, coagulation tests are useless for bleeding by 1.5 as compared to warfarin, whereas apix- level monitoring, albeit dabigatran will be considered to aban is similar to the latter (46). no longer have anticoagulant effect when APTT becomes normal (45). Using therapeutic doses the APTT value may be longer than the control by a factor of 2 or 3. WHAT INDICATIONS DO ANTICOAGULANTS In patients undergoing procedures with low bleeding HAVE? (TABLE VIII) risk discontinuation could be restricted to just one dose (that is, 10 h for dabigatran and apixaban; 20 h for ri- A gastroenterologist should know his trade, even more varoxaban) or could even be skipped for gastroscopy or so when indications for these drugs, which have reduced colonoscopy, with or without biopsy taking; in the pres- vascular mortality in patients at risk, are increasingly nu- ence of high risk (Table VII) they should be discontinued merous. for between 2 to 3 times the half-life (45). These drugs are Regarding atrial fibrillation, their indication is estab- reintroduced after 24 hrs as their fast action provides an- lished for patients with valve involvement and patients ticoagulation immediately. However, this may be delayed with thrombotic risk according to the CHA2DS2-VASc (47) for 2 days for procedures with high bleeding risk. classification (C- congestive heart failure or left ventric-

Table VII. NOAC management in high bleeding-risk procedures Dabigatran Rivaroxaban Apibaxan CrCl -5 -4 -3 -2 -1 0 +1 +2 +3 -5 -4 -3 -2 -1 0 +1 +2 +3 -5 -4 -3 -2 -1 0 +1 +2 +3 ≥ 80 Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes 50-80 Yes Yes No No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes 30-50 Yes No No No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes ≤ 30 n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. n.i. APTT Yes No No

Rev Esp Enferm Dig 2015; 107 (5): 289-306 298 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

Table VIII. Indications of oral anticoagulants Indication Recommended INR Duration Deep venous thrombosis 2-3 Postoperative or reversible risk factors > 3 months * Idiopathic: > 6 months Pulmonary embolism 2-3 * Relapsing or persistent cause: indefinite From three weeks before cardioversion to at 4 weeks Selective cardioversion 2-3 after cardioversion Biological valve prostheses or mitral valve disease (includes mitral prolapse) with some of the following: LA > 55 mm, 2-3 Indefinite thrombi inside, A-Fib, prior PTE Dilated cardiomyopathy with LVEF < 25% 2-3 Indefinite Extensive prior AMI with some of the following: 2-3 3 months intraventricular thrombi, dilated LV, LVEF < 35%, CHF Recurrent AMI 2.5-3.5 Indefinite Mechanical valve prostheses (Medtronic May, Indefinite: (if PTE during refer to Cardiology to consider 2.5-3.5 anticoagulation, Carbomedics, St Jude) adding antiplatelets) Mechanical valve protheses (older, Starr-Edwards type) 3-4.5 Indefinite Antiphospholipid syndrome 2.5-3.5 Indefinite

Atrial fibrillation, CHA2DS2-VASc ≥ 1 2-3 Indefinite ular dysfunction, H- hypertension, A- age ≥ 75, D- dia- – S- History of stroke (1 point), especially of the la- betes, S- prior stroke or thrombotic accident (CVA, TIA, cunar type. PTE), V- vascular disease (prior myocardial infarction, – B- History of bleeding, anemia, or predisposition to peripheral arteriopathy or aortic plaques), A- age between bleeding (1 point). 65 and 74, and S- female sex). Anticoagulation is deemed – L- Labile, unstable or high INR, or INR with less indicated when a criterion is met in the setting of atrial than 60% of time within the therapeutic range fibrillation (48) (Table IX). The higher the number of cri- (1 point). teria met, the higher the risk for thrombosis should antico- – E- Age ≥ 65 years (1 point). agulation be discontinued (49). – D- Drugs and/or alcohol: antiplatelet drugs (1 point), On the other hand, when making decisions for patients drinking 8 or more alcohol-containing beverages per with heart disease the HAS-BLED score is also used, week (1 point), or both (2 points). which assesses bleeding risk: A score of 3 or above suggests a higher bleeding risk – H- Hypertension: 1 point if not controlled, with SBP the year after anticoagulation, hence increased follow-up ≥ 160. or dose reductions for anticoagulants such as dabigatran – A- Impaired renal (1 point) or hepatic (1 point) or might be indicated. renal and hepatic (2 points) function.

IS ANTICOAGULATION A CONTRAINDICATION Table IX. Thrombotic risk in A-Fib FOR ANY ENDOSCOPIC TECHNIQUE? CHA DS -VASc score 2 2 – Yes, for endoscopic techniques with high bleeding Ref. Condition Points risk. C Congestive heart failure / LV dysfunction 1 – No, not for diagnostic procedures, including biopsy H Hypertension 1 taking (7). – In the emergency setting any technique may be con- A Age 75 yrs 2 ≥ sidered after reversing anticoagulation. D Diabetes 1 There is doubt about the adequate INR value for en- S Prior stroke or TIA or PTE 2 doscopic procedures with high bleeding risk. Endoscop- V Vascular disease (AMI, atheromatosis) 1 ic guidelines mention levels above 1.4 on the day of the procedure (laboratory monitoring is mandatory), although A Age = 65-74 yrs 1 some cardiology guidelines advocate for levels below 2.0. S Female sex 1 Bleeding risk, as already mentioned, continues to increase

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 299

even when guidelines are properly followed, thus we may – For polypectomy possibilities equal to or lower than assume the 1.4 value as general recommendation, even if 1%, continue with warfarin. higher levels might be endorsed in situations with very There is a somewhat defined agreement on the right high thrombotic risk, such as the use of metallic prostheses. regimen, according to both endoscopy type and anticoagulation indication (9). Attitude will be defined based on the endoscopic pro- WHICH CONDITIONS ENTAIL A HIGHER cedure’s bleeding risk, and then action will be according THROMBOTIC RISK WHEN REVERSING to the type of anticoagulant the patient is on: If the pa- ANTICOAGULATION? tient is taking acenocoumarol or warfarin (the latter must be discontinued for one extra day as compared to aceno- The ASGE (2) defines thrombotic risk as comprising coumarol) we must initiate a bridge therapy with LMWH two levels (Table X). adjusted to thrombotic risk. If on a NOAC, decisions This has been generally endorsed by other authors, in will be made depending on creatinine clearance (CrCl), manuals (50), reviews (51), and clinical guidelines (7), as the discontinuation scheme varies depending upon even though the CHA2DS2-VASc classification is fa- the molecule in use (Fig. 6). Whether a bridge therapy voured in cardiology environments, specifically when with LMWH should be used for patients previously on a referring to atrial fibrillation. Distinguishing three levels NOAC is subject to debate, but should discontinuation be of thrombotic risk might be practical when considering adjusted according to CrCl the patient will not be unpro- the various options available to act and to apply bridge tected or overmedicated. therapies (52) (Table XI), and this is what we include in The reintroduction of an anticoagulant after the proce- our proposed algorithm. We distinguish between high and dure will usually take place after 6 hours, except in high- very high risk mainly for practical purposes, since only risk situations, where it may be delayed 24 hours should in very high risk settings have the prophylactic benefits thrombotic risk allow it. Subjects on bridge therapy will of enoxaparin in divided doses, as opposed to other LM- be dosed at 24 hours, and LMWH may be extended and WHs, been demonstrated. anticoagulant reintroduction delayed when bleeding risk is high, always with agreement by a cardiologist, neurologist or hematologist. WHICH IS THE MOST PRACTICAL REGIMEN TO REVERSE ANTICOAGULATION? WHAT IS THE ROLE OF LOW MOLECULAR In a decision analysis based on the implementation of WEIGHT HEPARIN IN THE REVERSAL OF the ASGE clinical guidelines (2,4) the following strate- ANTICOAGULATION? gies were deemed most cost-effective (53): – In patients with low thrombotic risk (for instance In their guidelines, the ASGE (4) suggests a manage- atrial fibrillation without valve disease) or if pol- ment scheme using low molecular weight heparin as ypectomy possibilities exceed 60%, stop warfarin bridge for high thrombotic risk patients; the problem is 5 days before. that neither doses nor the proper timing for heparin rein- – In screening colonoscopies, where polyps are ex- troduction have been defined, although this is deemed to pected in over 35% of patients, maintain warfarin at oscillate between 2 and 6 hours after the procedure, and reduced dosage. consensus is to be sought with the remaining specialists

Table X. Thrombosis risk in patients on anticoagulants (ESGE) Low risk (1-2% if anticoagulation is interrupted for 4-7 days) High risk – Deep venous thrombosis – A-Fib with valve disease – Paroxysmal or chronic A-Fib not complicated with valve disease – Metallic mitral prosthesis – Biological valve prosthesis – Metallic valve with previous thrombotic event – Metallic aortic prosthesis

Table XI. Thrombosis risk during anticoagulation Low High Very high

A-Fib with CHA2DS2-VASc = 1-2 A-Fib with CHA2DS2-VASc > 3-5, PTE A-Fib with CHA2DS2-VASc > 5 Antiphospholipid syndrome Mechanical mitral or tricuspid prosthesis Previous multiple thromboses

Rev Esp Enferm Dig 2015; 107 (5): 289-306 300 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

ENDOSCOPY FOR ANTICOAGULATED PATIENTS ON VITAMIN K ANTAGONISTS

High bleeding-risk endoscopy Low bleeding-risk endoscopy

Undelayable condition Delayable condition Perform endoscopy

INR reversal < 1.4 Acenocoumarol/Warfarin NOAC – Vitamin K – Prothrombin complex Very high High Low Withdraw concentrate thrombotic risk thrombotic risk thrombotic risk according to CrCl data CA-AFib with CA-AFib with CA-AFib with

Perform endoscopy CHA2DS2-VASc >5, CHA2DS2-VASc 2-5, CHA2DS2-VASc 1-2, mechanical mitral PTE, mechanical aortic valve disease, Perform prosthesis valve with prior PTE, biologic mitral prosthesis endoscopy antiphospholipid sd., multiple prior thromboses

Discontinue Discontinue Discontinue anticoagulant anticoagulant anticoagulant Bridge therapy with Bridge therapy with Bridge therapy with LMWH LMWH (full dose) LMWH (half dose)

INR < 1.4* INR ≥ 1.4* *In very high bleeding-risk procedures (sphincterotomy, etc.) Perform Delay with INR < 1.4. For polypectomy, endoscopy endoscopy etc., 2.0 is enough

Fig. 6. Anticoagulation management schedule. involved. Other authors suggest a reduction of LMWH 3,500 units of bemiparin (57). However, this was about doses before interventional procedures, with early reintro- gastroscopy and colonoscopy with low bleeding risk, and duction in high-risk patients, even if this is described for a dose of 3,500 units might be inadequate in high throm- diagnostic catheterisms rather than techniques with high botic risk settings. bleeding risk in an assay (Table XII). However, bridge therapy poses some concern regard- LMWH has been used in high and double doses (54). ing effect enhancement when giving two anticoagulants However, there is no clear evidence regarding its supe- concomitantly, which may increase bleeding risk. Fur- riority over low, single doses, but bleeding does seem to thermore, renal LMWH clearance may be problematic increase (7). Many coagulation clinics establish dose ti- because of overdosing in subjects with uncontrolled or tration regimens according to thrombotic risk, and every unknown renal failure. site should define theirs in agreement with hematologists and cardiologists. The effectiveness and safety of various LMWH WHAT SHOULD BE DONE IN CASES WITH schemes as bridge therapy for the management of antico- CONCOMITANT ANTIAGGREGATION AND agulation withdrawal has been demonstrated in non-ran- ANTICOAGULATION? domized studies with enoxaparin sodium (55), dalteparin (56) or bemiparin (57), the latter providing the potential Such regimen would only be indicated for a condition benefit of single doses. A Spanish paper assesses its pro- demanding oral anticoagulation (atrial fibrillation, deep tective effect for gastroscopy and colonoscopy using venous thrombosis, pulmonary embolism, or valvular a user-friendly protocol consisting of single doses of prosthesis) and coronary disease (acute coronary syn-

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 301

Table XII. Bridge therapy regimen with anticoagulants -5 -4 -3 -2 -1 0 1 2 3 4 INR > 1.4 → Kovacs, LMWH LMWH Vit. K 1 mg LMWH 100/ HBPM 200/kg LMWH LMWH Disc OAC 2004 200/kg 200/kg INR < 1.4 → kg Init. OAC 200/kg 200/kg disc Zuckerman, Disc OAC LMWH LMWH LMWH withhold LMWH LMWH LMWH 2005 Constans, Disc Bemip Nothing Bemip 3500 2007 warfarin Init. AO Bemip 3500 Bemip 3500 Bemip 3500 Bemip 3500 Bemip 3500 Disc Bemip 3500 3500 Nothing acenocm drome and/or stent implantation) (58). At any rate, it is IS SCREENING COLONOSCOPY A TECHNIQUE advisable that this scheme be maintained for the short- REQUIRING A DIFFERENT BEHAVIOUR AS est time possible. No recommendations are available in COMPARED TO STANDARD DIAGNOSTIC the existing guidelines and papers in endoscopic pro- COLONOSCOPY? cedures. Recommendations do exist in some consensus European guidelines in the field of cardiology, where it In the various series of screening colonoscopies in our is stated that in the presence of severe acute bleeding setting, the possibility of significant polypectomy –for and triple therapy the monitoring of INR levels below polyps larger than 10 mm– is encountered in 50% to 60% 2 may be necessary while maintaining clopidogrel and of patients. low-dose aspirin. This is an increasingly common sit- As seen above, a reduction of anticoagulant doses is con- uation, and cases should be managed on an individual sidered to be cost-effective (50), albeit this is uncommon in basis. our setting. In these patients a bridge therapy with LMWH is perhaps most appropriate from a practical standpoint. As regards antiplatelets, aspirin will not be withdrawn, WHAT SHOULD BE DONE FOR PATIENTS regardless of dosage. If on clopidogrel, we shall inform ON ASPIRIN, ANTIPLATELETS OR the patient that the bleeding risk is slightly increased, or ANTICOAGULANTS OUTSIDE ACCEPTED we may change it for aspirin, and about the possibility of INDICATIONS? delaying the procedure until it may be withdrawn. If on dual antiaggregation, consider clopidogrel dis- In such situations therapy discontinuation before the continuation if possible (as per established schemes) and endoscopic procedure would be most appropriate, as in- maintain aspirin. creasing the risk of bleeding, however small, while main- With a consensual attitude in our team, we chose to taining the thrombotic risk would be senseless. manage screening as a high-risk endoscopy given the high rate of polypectomies (> 50%), in order to prevent second procedures from happening, although patients should be MAY WE RECOMMEND A PRACTICAL involved in the decision-making. SCHEME FOR ANTIAGGREGATED AND/OR ANTICOAGULATED PATIENTS? HOW IMPORTANT IS PATIENT PERSPECTIVE Mixed management together with cardiologists, hema- WHEN THE DECISION IS MADE TO tologists, and neurologists is important, considering pa- DISCONTINUE AN ANTICOAGULANT OR tient conditions on an individual basis. ANTIPLATELETS DRUG? While a practical proposal is made at the end of the protocol (Annex 1), we must combine the risk estimates This point was discussed given the relevance of dis- and feelings the endoscopist may have after the procedure continuation for patient health, and patients prefer the risk with those of a cardiologist, hematologist or neurologist of side effects (including bleeding) to the risk of poten- in order to agree on the timing for discontinuation (rather tial cardiovascular events (59). This appreciation should stable) and reintroduction (more flexible). prompt us to make decisions on an individual basis prior Guidelines are helpful in the decision-making under- to the procedure, with time enough for the patient to make lying medical acts, rather than tight schemes to be fol- up his or her mind. lowed stringently, as evidence for the latter is presently The finding of polyps during colonoscopy is increas- weak. ingly common in patients on anticoagulants and/or anti-

Rev Esp Enferm Dig 2015; 107 (5): 289-306 302 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

platelet drugs. In such case decisions should be agreed 15. Heidbuchel H, Verhamme P, Alings M, et al. European Heart Rhythm upon before the procedure, particularly for sedated pa- Association Practical Guide on the use of new oral anticoagulants in patients with non-valvular atrial fibrillation. Eurospace 2013;15: tients. The informed consent should approach potential 625-51. polypectomy, and describe its benefits and risks, as well 16. Parra-Blanco A, Kaminaga N, Kojima Y, et al. Hemoclipping for pos- as the need to repeat both the procedure and the prepara- tpolypectomy and postbiopsy colonia bleeding. Gastrointest Endosc 2000;51:37-41. tion should polypectomy be decided against. 17. Smith LE. Fiberoptic colonoscopy: Complications of colonoscopy and polyupectomy. Dis Colon Rectum 1976;19:407-12. 18. Sieg A, Hachmoeller-Eisenbach U, Eisenbach T. Prospective evalua- ACKNOWLEDGEMENTS tion of complications in outpatient GI endoscopy: A survey among German gastroenterologists. Gastrointest Endosc 2001;53:620-7. 19. Di Giorgio P, De Luca L, Calcagno G, et al. Detachable snare versus We are grateful to the board members of SEED, SEPD, epinephrine injection in the prevention of postpolypectomy bleeding: GTCV-SEC, and SETH for their contributions and correc- A randomized and controlled study. Endoscopy 2004;36:860-3. 20. Bason MD, Manzini L, Palmer RH. Effect of nabumetone and aspi- tions in order to improve and make sense of this document. rin on colonic mucosal bleeding time. Aliment Pharmacol Ther We particularly acknowledge Dra. Inmaculada Roldán 2001;15:539-42. (Servicio de Cardiología, Hospital La Paz, Madrid) and 21. Nakajima H, Takami H, Yamagata K, et al. Aspirin effects on colonia Dr. Enric Brullet (Servicio de Digestivo, Hospital Parc mucosal bleeding: implications for colonia biopsy and polypectomy. Dis Colon Rectum 1997;40:1484-8. Taulí, Sabadell) for their specific contributions. 22. Hui AJ, Wong RM, China JK, et al. Risk of colonoscopic polipectomy bleeding with anticoagulants and antiplatelet agents: analisis of 1657 cases. Gastrointestinal Endoscopy 2004. REFERENCES 23. Shiffman ML, Farrel MT, Yee YS. Risk of bleeding alter endoscopic biopsy or polypectomy in patients taking aspirin or other NSAIDs. Gastrointestinal Endoscopy 1994;40:458-62. 1. Ezekowitz MD. Anticoagulation interruptus: Not without risk. Circu- 24. Yousfi M, Gostout CJ, Baron TH, et al. Postpolypectomy lower gas- lation 2004;110:1518-9. trointestinal bleeding: Potencial role of aspirin. Am J Gastroenterol 2. Eisen, GM, Baron, TH, Dominitz, JA, et al. Guideline on the manage- 2004;99:1785-9. ment of anticoagulation and antiplatelet therapy for endoscopic pro- 25. Kimchi NA, Broide E, Scapa E, et al. Antiplatelet therapy and the cedures. Gastrointest Endosc 2002;55:775. risk of bleeding induced by gastrointestinal endoscopio procedures. A 3. Baker RI, Coaughlin PB, Gallus AS, et al.; the Warfarin Reversal Con- systematic review of the literature and recommendations. Digestion sensos Group. Warfarin reversal: Consensus guidelines, on behalf of 2007;75:36-45. the Australasian Society of Trombosis and Haemostasis. Medical Jour- 26. Witt DM, Delate T, McCool KH, et al.; WARPED Consortium. Inci- nal of Australia 2004;181:492-7. dence and predictors of bleeding or thrombosis after polypectomy in 4. Zuckerman, MJ, Hirota, WK, Adler, DG, et al. ASGE guideline: The patients receivin g and not receiving anticoagulation therapy. J Thromb management of low-molecular-weight heparin and nonaspirin anti- Haemost 2009;7:1982-9. platelet agents for endoscopic procedures. Gastrointest Endosc 2005; 27. Gerson LB, Michaels L, Ullah N, et al. Adverse events associated 61:189-94. with anticoagulation therapy in the periendoscopic period. Gastrointest 5. Boustière C, Veitch A, Vanbiervliet G, et al. ESGE Guideline: Endos- Endosc 2010;71:1211-7. copy and antiplatelet agents. Endoscopy 2011;43:445-58. 28. Friedland S, Sedehi D, Soetikno R. Colonoscopic polypectomy in anti- 6. Alberca de las Parras F. Guía de práctica clínica para el manejo de la coagulated patients. World J Gastroenterol 2009;15:1973-6. coagulación en pacientes sometidos a técnicas endoscópicas. Rev Esp 29. Hirouchi A, Nakayama Y, Kajjyama M, et al. Removal of small colo- Enferm Dig 2010;102:124-38. rectal polyps in anticoagulated patients: a prospective randomized 7. Veitch AM, Baglin TP, Gershlick AH, et al. Guidelines for the manage- comparison of cold snare and conventional polypectomy. Gastrointest ment of anticoagulant endoscopic procedures and antiplatelet therapy Endosc 2013;79:417-2. in patients undergoing endoscopic procedures. Gut 2008;57:1322-9. 30. Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy 8. Becker RC, Scheiman J, Dauerman HL, et al; American College of complications and their management: An attemp at consensus. Gas- Cardiology and the American College of Gastroenterology. Mana- torintest Endosc 1991;37:383-93. gement of platelet-directed pharmacotherapy in patients with athe- 31. Hui CK, Lai KC, Yuen MF, et al. Does withholding aspirrin for one rosclerotic coronary artery disease undergoing elective endoscopic week reduce the risk of post-sphinsterotomy bleeding? Aliment Phar- gastrointestinal procedures. Am J Gastroenterol 2009;104:2903-17. macol Ther 2002;16:929-36. 9. Gerson LB, Gage BF, Owens DK, et al. Effect and outcomes of the 32. Nelson DB, Freeman ML. Mayor haemorrhage from endoscopio ASGE guidelines on the periendoscopic management of patients who sphincterotomy: risk factor analysis. J Clin Gastroenterol 1994;19: take anticoagulants. Am J Gastroenterol 2000;95:1717-24. 283-7. 10. Fujishiro M, Oda I, Yamamoto Y, et al. Multi-center survey regarding 33. Weinberg BM, Shindy W, Lo S. Dilatación esfinteriana endoscópica the management of anticoagulation and antiplatelet therapy for endos- con balón (esfinteroplastia) versus esfinterotomía para los cálculos del copic procedures in Japan. J Gastroenterol Hepatol 2009;24:214-8. conducto biliar común (Revisión Cochrane traducida). En: La Biblio- 11. Radaelli F, Paggi S, Terruzzi V, et al. Management of warfarin-asso- teca Cochrane Plus, 2007 Número 4. Oxford: Update Software Ltd. ciated coagulopathy in patients with acute gastrointestinal bleeding: Disponible en: http://www.update-software.com. (Traducida de The A cross-sectional physician survey of current practice. Dig Liver Dis Cochrane Library, 2007 Issue 4. Chichester, UK: John Wiley & Sons, 2011;43:444-7. Ltd.). 12. Ono S, Fujishiro M, Hirano K, et al. Retrospective analysis on the 34. Mönkemüller K, Fry LC, Neumann H, et al. Diagnostic and therapeutic management of anticoagulants and antiplatelet agents for scheduled utility of double balloon endoscopy: experience with 225 procedures. endoscopy. J Gastroenterol 2009;44:1185-9. Acta Gastroenterol Latinoam 2007;37:216-23. 13. Lobo B, Saperas E. Tratamiento de la hemorragia digestiva por rup- 35. Koh R, Hirasawa K, Yahara S, et al. Antithrombotic drugs are risk tura de varices esofágicas. http://www.prous.com/digest/protocolos/ factors for delayed postoperative bleeding after endoscopic submu- view_protocolo.asp?id_protocolo=18. 2004 Prous Ed. cosal dissection for gastric neoplasms. Gastrointestinal Endoscopy 14. Cukor B, Cryer BL. The risk of rebleeding after an index GI bleed in 2013;78:476-83. patients on anticoagulation (abstract DDW 2008). Gastrointest Endosc 36. Singh D, Laya AS, Vaidya OU, et al. Risk of bleeding after percuta- 2008;67:AB24. neous endoscopic gastrostomy (PEG). Dig Dis Sci 2012;57:973-80.

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 303

37. Camm AJ, Bassand JP, Agewall S, et al. ESC Guidelines for the mana- using a novel risk factor-based approach: The Euro Heart Survey on gement of acute coronary syndromes in patients presenting without Atrial Fibrillation. Chest 2010;137:263-72. persistent ST-segment elevation. The Task Force for the management 48. Camm AJ, Lip GYH, De Caterina R, et al. 2012 focused update of the of acute coronary syndromes (ACS) in patients presenting without ESC Guidelines for the management of atrial fibrillation. An update persistent ST-segment elevation of the European Society of Cardiology of the 2010 ESC Guidelines for the management of atrial fibrillation. (ESC). European Heart Journal 2011;32:2999-3054. European Heart Journal 2012;33:2719-47. 38. Steg G, James SK, Atar D, et al. ESC Guidelines for the management 49. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical clas- of acute myocardial infarction in patients presenting with ST-segment sification schemes for predicting stroke: Results from the National elevation. The Task Force on the management of ST-segment elevation Registry of Atrial Fibrillation. JAMA 2001;285:2864-70. acute myocardial infarction of the European Society of Cardiology 50. Morillas JD, Simón MA. Endoscopia: preparación, prevención y tra- (ESC) European Heart Journal 2012;33:2569-619. tamiento de las complicaciones. En: Tratamiento de las enfermedades 39. Fuentes B, Gállego J, Gil-Nunez A, et al.; por el Comité ad hoc del gastroenterológicas. Asociación Española de Gastroenterología. Ed Grupo de Estudio de Enfermedades Cerebrovasculares de la SEN. Guía SCM; 2006. para el tratamiento preventivo del ictus isquémico y AIT (II). Reco- 51. Kamath PS. Gastroenterologic procedures in patients with disorders of mendaciones según subtipo etiológico. Neurología 2014;29:168-83. hemostasis. Up To Date. http://www.uptodate.com/physicians/gasthe- 40. Grines CL, Bonow RO, Casey DE Jr., et al. Prevention of premature pa_toclist.asp 2007. discontinuation of dual antiplatelet therapy in patients with coronary 52. You JJ, Singer DE, Howard PA, et al. Antithrombotic Therapy for artery stents: A science advisory from the American Heart Associa- Atrial Fibrillation: Antithrombotic Therapy and Prevention of Throm- tion, American College of Cardiology, Society for Cardiovascular bosis. 9th ed. American College of Chest Physicians Evidence-Based Angiography and Interventions, American College of Surgeons, and Clinical Practice Guidelines. Chest 2012;141(2 Supl.):e531S-e575S. American Dental Association, with representation from the American doi:10.1378/chest.11-2304 College of Physicians. Circulation 2007;115:813-8. 53. Gerson LB, Triadafilopoulos G, Gage BF. The management of anticoa- 41. Fleisher LA, Beckman JA, Brown KA, et al. Guidelines on perioperati- gulants in the periendoscopic period for patients with atrial fibrillation: ve cardiovascular evaluation and care for noncardiac surgery: A report a decision analysis. Am J Med 2004;116: 451-9. of the American College of Cardiology Foundation/American Heart 54. Seshadri N, Goldhaber SZ, Elkayam U, et al. The clinical challen- Association Task Force on Practice Guidelines. J Am Coll Cardiol ge of brinding anticoagulation with low-molecular-weigth heparin in 2007;50;159-241. patients with mecanical prosthetic heart valves: an evidence-based 42. Cohen Arazi H, Carnevalini M, Falconi E, et al. The association of comparative review focusing on anticoagulation options in pregnant antiplatelet aggregation effect of aspirin and platelet count. Possible and nonpregnant patients. Am Heart J 2005;150:27-34. dosage implications. Revista Argentina de Cardiología 2012;80. 55. Douketis JD, Johnson JA, Turpie AG. Low-molecular-weith heparin 43. Fox KA, Mehta SR, Peters R, et al.; Clopidogrel in Unstable angina as bringing interruption of warfarin. Arch Intern Med 2004;164:1319- to prevent Recurrent ischemic Events Trial. Benefits and risks of the 26. combination of clopidogrel and aspirin in patients undergoing surgical 56. Turpie AGG, Johnson J. Temporary discontinuation of oral anticoagu- revascularization for non-ST-elevation acute coronary syndrome: the lants: Role of low molecular weight heparin (dalteparin). Circuation clopidogrel in unestable angina to prevent recurrent ischemic events 200;102:II-826. (CURE) trial. Circulation 2004;110:1202. 57. Constans M, Santamaria A, Mateo J, et al. Low-molecular-weight 44. Savonitto S, Urbano MD, Caracciolo M, et al. Urgent surgery in patients heparin as bridging therapy during interruption of oral anticoagula- with a recently implanted coronary drug-eluting stent: a phase II study tion in patients undergoing colonoscopy or gastroscopy. Int Clin J of “bridging” antiplatelet therapy with tirofiban during temporary with- Prac 2007;61:212-7. drawal of clopidogrel. British Journal of Anaesthesia 2010;104:285-91. 58. Wijns W, Kolh P, Danchin N, et al. Guidelines on myocardial revas- 45. Desai J, Granger CB, Weitz JI, et al. Novel oral anticoagulants in gas- cularization. The Task Force on Myocardial Revascularization of the troenterology practice. Gastrointestinal Endoscopy 2013;78:227-39. European Society of Cardiology (ESC) and the European Associa- 46. Desai J, Kolb JM, Weitz JI, et al. Gastrointestinal bleeding with the tion for Cardio-Thoracic Surgery (EACTS) European Heart Journal oral anticoagulants- defining the issues and the management strategies. 2010;31:2501-55. Thrombosis and Haemostasis 2013;110:205-12. 59. Devereaux PJ, Anderson DR, Gardner MJ, et al. Differences between 47. Lip GY, Nieuwlaat R, Pisters R, et al. Refining clinical risk stratifica- perspectives of physicians and patients on anticoagulation in patients tion for predicting stroke and thromboembolism in atrial fibrillation with atrial fibrillation: observational study. BMJ 2001;323:1218-22.

Rev Esp Enferm Dig 2015; 107 (5): 289-306 304 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

ANNEX 1

PATIENT INFORMATION

You are to undergo a ______procedure on ______, as requested by Dr. ______, at the ______clinic, with diagnostic or therapeutic intent.

Since you are taking the following medications ______, your bleeding risk is increased, hence adjusting your doses may be desirable in order to reduce this risk.

Furthermore, you are taking these drugs to prevent thrombosis, hence any adjustments should be assessed by your doctor. This is why this management protocol was developed, based on the available medical literature. Any application of this protocol is to be attempted under medical surveillance and judgment, and schemes may be changed according to clinical variability criteria. If pregnant or at risk of pregnancy, please inform your doctor.

It is advisable for patients previously with poorly controlled anticoagulation that adjustments be made at the Anticoagulation Room

In cardiology patients with difficult decision-making because of intermediate situations, refer to cardiologist to reach agreement on ensuing actions.

The requesting doctor is to mark every option appropriate for his/her patient

TREATMENT WITH ANTIPLATELETS

1. Techniques with LOW BLEEDING RISK are to be performed (gastroscopies/colonoscopies with/without biopsy, single polipectomies, echoendoscopy without puncturing, prosthesis placement without dilation): □ Drug discontinuation and dose adjustments are unnecessary. Biopsies may be taken from moderately vascularized lesions. □ Should lesions amenable to treatment (such as polyps) be found during the procedure, they may be excised if the endoscopist judges bleeding risk is low.

2. Techniques with HIGH BLEEDING RISK are to be performed (complex polypectomy, sphincterotomy and ERCP, dilation, endoscopic gastrostomy, echoendoscopy with puncturing, varices management, balloon enteroscopy, liver biopsy; also screening colonoscopy should the patient consent).

2.1. Situations with very low thrombotic risk (primary prevention). □ Antiplatelet may be discontinued.

2.2. Situations with low thrombotic risk [uncovered stents (> 6 months) and covered stents (> 18 months)]: Patients are usually on aspirin alone (ASA®, Adiro®, Bioplak®, Therasacard®, Tromalyt®) or triflusal (Anpeval®, Disgren®): □ No dose adjustment necessary, may be performed while on these drugs. □ Should a particularly high bleeding risk exist, these may be discontinued and the procedure may be scheduled in their absence.

2.3. Situations with intermediate thrombotic risk [uncovered stents (1-6 months) and covered stents (12-18 months), acute coronary syndrome with stent (6-12 months)]: they are usually on monotherapy with ticlopidine (Tiklid®) or clopidogrel (Plavix®, Iscover®), or some times on dual antiplatelet therapy: □ Replace antiplatelet______with aspirin, 100 mg/day. □ In case of dual antiaggregation, discontinue clopidogrel 5 days before, and perform with aspirin. □ Attempt to delay endoscopy until stent-related thrombotic risk is lower (go to section 2.2).

2.4. Situations with high thrombotic risk [uncovered stents (1 month) and covered stents (12 months), acute coronary syndrome without stent (12 months) and with stent (1-6 months)]: they are usually on dual therapy [ASA + clopidogrel (or prasugrel (Effient®) or ticagrelor (Brilique®)]:

Rev Esp Enferm Dig 2015; 107 (5): 289-306 Vol. 107, N.º 5, 2015 MANAGEMENT OF ANTITHROMBOTIC DRUGS IN ASSOCIATION WITH ENDOSCOPIC PROCEDURES 305

□ Attempt delaying endoscopy until thrombotic risk is reduced (consult with cardiology) and perform the procedure 5 days following ticlopidine or clopidogrel discontinuation, or 7 days following ticagrelor or prasugrel discontinuation, keeping aspirin though. □ If the procedure cannot be delayed, the use of tirofibran as bridge therapy may be considered, following consultation with cardiology, as it requires admission and IV medication.

Antiplatelets will be reintroduced the next day after the procedure, except when the endoscopist feels there is a high risk of bleeding - in such a case, discuss with cardiology/neurology/hematology.

FILL IN PATIENT INSTRUCTIONS (circle corresponding options)

Day -7 -6 -5 -4 -3 -2 -1 Test +1 +2 +3 Date Antiplatelet 1 ______Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes Yes Yes Antiplatelet 2 ______Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No No Yes Yes Yes

TREATMENT WITH ANTICOAGULANTS

1. Endoscopies with LOW BLEEDING RISK are to be performed (gastroscopies/colonoscopies with/without biopsy, single polipectomies, echoendoscopy without puncturing, prosthesis placement without dilation): □ Endoscopy, even biopsy taking, may be performed without risk. □ However, to be safe, the scheme 2 may be performed in the corresponding group, just in case a polyp or another manageable lesion is found during the procedure, and a future repeat procedure is undesirable.

2. Endoscopies with HIGH BLEEDING RISK are to be performed:

2-1. Patients on acenocoumarol (Sintrom®) or warfarin (Aldocumar®)

Day -4 -3 -2 -1 0 +1 +2 +3 +4 Date Acenocumarol yes no no no PD PD PD PD PD Warfarin no Very high (CA x AFib with Ac. no no CHA DS -VASc > 5, mitral 2 2 ENOXAPARIN no yes yes* no yes* yes yes or tricuspideal mechanical Wf. yes yes prosthesis) BEMIPARIN Ac. no no High (CA x AFib with DALTEPARIN CHA DS -VASc 3-5, prior PTE, 2 2 ENOXAPARIN no yes yes* no sí* sí yes antiphospholipid syndrome, Wf. yes yes NADROPARIN prior multiple thrombosis) THROMBOSIS RISK TINZAPARIN Low (CA x AFib with No bridge therapy CHA2DS2-VASc 1-2) Testing previously (*): INR < 1,4 X

Rev Esp Enferm Dig 2015; 107 (5): 289-306 306 F. ALBERCA-DE-LAS-PARRAS ET AL. Rev Esp Enferm Dig (Madrid)

LMWH Prophylactic dose Therapeutic dose CHA2DS2-VASc BEMIPARIN (Hibor©) 2500-3500 U/d 115 U/kg/d C-Heart failure/dysfunction VI 1 200 U/kg/d H-Hypertension 1 DALTEPARIN (Fragmin©) 2500-5000 U/d (o 100 U/kg/12 h) A-Age ≥75 yrs. 2 ENOXAPARIN (Clexane©) 20-40 U/d 1-1,5 mg/kg/12 h D-Diabetes 1 NADROPARIN (Fraxiparina©) 7500-15000 U/d 0,1 ml/10 kg/12 h S-Prior thrombosis (CVA,TIA, PTE) 2 TINZAPARIN (Innohep©) 3500-4500 U/d 175 U/kg/d V-Vascular disease (AMI, atheromatosis) 1 A-Age 65-74 1 S-Female sex 1

- The last dose of heparin will be 24 hours before the test (adjusted if cited in the afternoon). - To test it takes INR below 1.4 - DP: Previous dose. On the day of the test will be as soon as 6 hours after it. If you consider that there is a high risk of bleeding can prolong and delay the restart LMWH anticoagulant. - The dose of LMWH will be considered therapeutic (see box).

b/ Patients with dabigatran (Pradaxa®), rivaroxaban (Xarelto®) or apixaban (Eliquis®):

DABIGATRAN RIVAROXABAN APIXABAN CrCl -5 -4 -3 -2 -1 0 +1 +2 +3 -5 -4 -3 -2 -1 0 +1 +2 +3 -5 -4 -3 -2 -1 0 +1 +2 +3 ≥ 80 Y Y Y N N N Y Y Y Y Y Y N N N Y Y Y Y Y Y N N N Y Y Y 50-80 Y Y N N N N Y Y Y Y Y Y N N N Y Y Y Y Y Y N N N Y Y Y 30-50 Y N N N N N Y Y Y Y Y Y N N N Y Y Y Y Y Y N N N Y Y Y ≤ 30 Not indicated Not indicated Not indicated ATTP yes no no

CrCl: creatinin clearance. Not indicated (may be used without >15 according to some guidelines) * Except for very high risk of bleeding as judged by the endoscopist - No reliable data available for edoxaban (not marketed) - ATTP will be assessed for patients with renal failure and Cr Cl < 50: if APTT >2 times above control, delay procedure - In patients with CrCl ≥ 80 and manageable bleeding risk they could be withdrawn for only 24 hours

TREATMENT WITH ANTICOAGULANTS + ANTIPLATELETS

The most appropriate scheme remains unclear, hence procedures with high bleeding risk should be delayed as much as possible, and true thrombotic risk should be evaluated jointly with the Cardiology, Neurology or Hematology department.

IDENTIFICATION OF PROBLEMS AFTER THE PROCEDURE

Should any suspected bleeding episode arise (blood in vomit, foul-smelling tarry stools or sustained dizziness with hypotension), please visit the emergency room taking this sheet and your test results with you.

Date: ______Signed: Dr. ______

For any questions or concerns, please contact the Digestive Endoscopy Unit (Tel.: 0)

Rev Esp Enferm Dig 2015; 107 (5): 289-306