Guidelines at a glance: the new ACCP guidelines on antithrombotic therapy (9th edition)

The title to this post is a little deceptive. There's way too much content here to take in at a glance. In fact this post may go on record as the longest one in history of The Notes. Nevertheless it's my best effort to distill the contents of this vast Chest Supplement down to the nuts and bolts of what hospitalists need to know. The first reference point in the document, and the one from which most of the content herein is cited, is the Executive Summary.


The creators of the guideline, purporting to make it more evidence based, cite in their introduction a diminished role of thrombosis experts in the 9th edition:

That solution is to give primary leadership and responsibility for each article not to a thrombosis expert but to a methodologist who, in almost all cases, also is a practicing physician without important conflicts of interest4,5 ..

They correctly point out that not only financial but also academic conflicts of interest are problematic for experts. But favoring methodologists over experts to lead guideline development is disturbing. Here's the definition of EBM from the landmark article by Sackett and others (my italics):

Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.

In order to judiciously apply evidence doctors need more than a recitation and critical appraisal of published literature when they refer to guidelines. This is where expertise is important. Expertise adds depth. (For a thoroughly conflicted and superb review of superficial thrombophlebitis and an article that belongs in every hospitalist's library see here).


The new guidelines place greater emphasis on patient preferences and values than prior editions. It is not uncommon in the guidelines, for example, to find statements such as this:

Remarks: Patients who are more concerned about avoiding the unknown, but potentially large increase in bleeding risk associated with the perioperative continuation of dual antiplatelet therapy than avoiding the risk for coronary stent thrombosis are unlikely to choose continuation of dual antiplatelet therapy.

This reflects one of the founding principles of EBM: the individual patient is a key element in the process. It's by definition. Proponents of more central control of medical decision making, when they cite EBM with all its popular appeal (and misunderstanding) to bolster their arguments, are merely expressing their ignorance of what EBM really is.


But I digress. Following are some comments about changes from the previous edition of the guidelines along with points I think are of special interest to hospitalists:


Use of warfarin
Recommendations for the management of over anticoagulation with warfarin are no longer included in the executive summary of the guidelines, but are found in the body of this paper from the guideline supplement. For asymptomatic INRs up to and including 10 the authors suggest no active reversal. For asymptomatic INRs over 10 the authors suggest oral vitamin K. For major bleeding at any elevated INR, in a change from previous guidelines, the authors suggest reversal with 4 factor prothrombin complex concentrate (PCC) rather than plasma and that, in order to sustain the reversal effect, this be supplemented with IV vitamin K 5-10 mg via slow infusion. Unfortunately, as of the publication of this recent review, 4 factor PCC was not available in the US. The authors of that review discuss alternatives, such as the use of 3 factor PCC supplemented with small doses of recombinant factor VIIa or the use of factor eight inhibitor bypass activity.


The guideline recommends against routine pharmacogenetic testing of patients starting warfarin therapy.


The guideline suggests that reliable patients with stable INRs on warfarin can have their interval of INR testing extended to three months.


Concerning interacting drugs antibiotics have been controversial. The best evidence for adverse effects is concerning quinolones and co-trimoxazole. The authors suggest avoiding these drugs in patients taking warfarin. This may not apply in the inpatient setting where the INR can be monitored daily.


Outpatient use of unfractionated heparin
Patients eligible for outpatient VTE treatment may be treated with weight adjusted UFH without PTT monitoring, with outcomes comparable to LMH, based on evidence from one RCT.


VTE prophylaxis
Concerning VTE prophylaxis for hospitalized medical patients the ACCP is pretty much in line with the recently released ACP guidelines which restrict pharmacologic prophylaxis in medical patients to those at high risk for VTE and decry prophylaxis as a default option for all medical patients, as is now driven by some performance initiatives. The ACCP guidelines consider two groups of medical patients: critical and non critical. Critical patients, which for practical purposes could be defined as those in the ICU, are all considered high risk and deserving of pharmacologic prophylaxis provided there is not a high bleeding risk. Non-critical (ward) patients should be assessed by the Padua prediction score to determine whether they are at high risk for VTE.


Extending the course of VTE prophylaxis to the post discharge period has garnered a great deal of recent discussion. For medical patients the ACCP guidelines recommend against this practice. They also recommend stopping pharmacologic prophylaxis once the patient is mobile---which would seem to mean that prophylaxis can be stopped at discharge or as soon as the patient becomes ambulatory (anything more than bathroom privileges), whichever happens first.


For outpatients with solid tumors the guidelines suggest pharmacologic prophylaxis but only if the patient has additional risk factors (beyond the cancer) for VTE.


For long distance travelers the guidelines do not recommend pharmacologic prophylaxis although for those at high risk certain non-pharmacologic measures are suggested.

The VTE prevention guidelines for abdominal and pelvic surgical patients are complex and based on risk assessment via the Rogers and Caprini scores. Those tools are contained in the body of this paper from the supplement. If the surgery is for cancer, pharmacologic prophylaxis extended through 4 weeks is highly recommended. Recommendations based on the scores: Rogers 10 or less, Caprini 2 or less, no pharm prophylaxis. Pharm prophylaxis is recommended for higher scores.


Pharmacologic prophylaxis is not recommended for cardiac surgery patients without complications. It recommended for those whose course is prolonged by non-hemorrhagic complications.


Pharmacologic prophylaxis is suggested for non-cardiac thoracic surgery patients judged to be at moderate or high risk.


In craniotomy patients the guidelines suggest adding pharmacologic prophylaxis only in patients at very high risk and only after hemostasis is achieved and bleeding risk is decreased. Similar recommendations are applied to spinal surgery and trauma patients.


For orthopedic surgery patients total hip arthroplasty, total knee arthroplasty and hip fracture surgery may, for practical purposes, be considered as one group. Aspirin, strangely enough, is listed among the alternatives for pharmacologic prophylaxis. However low molecular weight heparin is favored over all the alternatives including fondaparinux, unfractionated heparin, warfarin and novel oral anticoagulants. Duration of prophylaxis is for a minimum of 10 days. Extension of prophylaxis for up to 35 days is suggested. It is suggested that mechanical prophylaxis be combined with pharmacologic prophylaxis. Doses for LMWH are not given in the guideline and the product labeling should be consulted. Recommended start time for pharmacologic prophylaxis is 12 or more hours before or after surgery.


No pharmacologic prophylaxis is recommended for distal leg surgery or knee arthroscopy.


A prophylactic IVC filter is NOT recommended in ANY surgical or trauma scenario, even when pharmacologic prophylaxis is contraindicated.


Perioperative management of patients on warfarin, bridging
Patients on warfarin who are scheduled for surgery should have it interrupted approximately 5 days before surgery and resumed 12 to 24 hours after, if hemostasis is achieved.


Bridging indications: whether in patients with mechanical heart valves, atrial fibrillation or VTE as the warfarin indication the decision to bridge or not depends on risk assessment. Yes for high, no for low, and individualized decision making for moderate. How is that risk determined? For mechanical heart valve prostheses mitral position, tilting disc or caged ball, or cerebrovascular event in the past 6 months constitute high; bileaflet aortic position with a fib, prior cerebrovascular event, HT, DM, CHF or age over 75 constitute moderate; aortic bileaflet absent the above constitutes low. For a fib, CHADS 2 of 5 or above, cerebrovascular event within 3 months or rheumatic disease constitute high; CHADS 2 of 3 or 4 constitutes moderate; CHADS 2 of 2 or less with no cerebrovascular history constitutes low. For VTE event within 3 months or severe thrombophilia (meaning protein C, S, or antithrombin deficiency or APLS) constitute high risk; non-severe thrombophilia, history of multiple VTEs, event within 12 months or active cancer constitute moderate; event greater then 12 months out absent the above constitutes low.


For minor procedures (dental, derm, cataract) stopping warfarin is not recommended across the board. For dental procedures, however, administration of a local oral hemostatic agent or interruption of warfarin for 2-3 days prior is suggested. Aspirin interruption is not recommended in any of these scenarios.


Perioperative management of patients on antiplatelet agents
It is suggested that ASA be continued perioperatively in non-cardiac surgery in patients at high cardiovascular risk, but interrupted 7-10 days prior if not.


For CABG, ASA continuation is recommended but stopping clopidogrel or congeners 5 days before.


For patients with coronary stents the guidelines recommend postponing surgery for at least 6 weeks post bare-metal stent placement and at least 6 months (as opposed to a year) following DES placement. If surgery cannot be delayed the guidelines recommend continuing dual antiplatelet therapy through the perioperative period rather than stopping them.


DVT treatments, acute
For isolated distal DVT serial imaging and no anticoagulation is favored if no severe symptoms or risk factors for extension. If the latter exist or extension (even still within the confines of the distal veins) is demonstrated then anticoagulation is recommended.


In the treatment of VTE parenteral anticoagulants are to be continued 5 days or until the INR is therapeutic for at least 24 hours whichever is longer.


For acute DVT all things being equal LMWH or fondaparinux is favored over UFH. The guideline writers recognize the convenience advantage of once over twice daily administration of LMWH when treating acute VTE with full therapeutic anticoagulation. However such once daily administration is recommended only for those LMWH products whose labeling calls for the same total daily dose in the once and twice daily regimens. In other words the labeling doubles the dose per injection for the once daily regimen. Thus enoxaparin, whose labeling for once daily dosing calls for 1.5 time the injection dose for twice daily use, would not be appropriate for once daily administration in the treatment of VTE.


Home over hospital treatment of acute DVT is recommended when circumstances are favorable. (This is not the case for PE---see below).


Anticoagulation alone is suggested over regional or systemic thrombolysis, or surgical clot removal. The writers acknowledge that, particularly in circumstances of high risk for PTS, patient values and preferences might favor one of the alternatives.


In patients with acute DVT the only recommended indication for an IVC filter is those patients with contraindication to anticoagulants.


DVT, anticoagulation duration
The authors do not consider the presence of a permanent IVC filter, by itself, as an indication for indefinite anticoagulation.


In terms of duration of anticoagulation for DVT the authors have defined the terms in a way that differs from prior popular usage:

In this review, the term long-term treatment refers to treatments (eg, VKA therapy, LMWH, dabigatran) that are continued after initial therapy (eg, parenteral anticoagulation, thrombolytic therapy) (Fig 1). In addition, we consider treatment with rivaroxaban, which is used without initial parenteral therapy. Long-term therapy has two goals: (1) to complete treatment of the acute episode of VTE and (2) to prevent new episodes of VTE that are not directly related to the acute event. During the early phase of long-term treatment (ie, first 3 months), treatment of the acute episode of VTE predominates. During the late phase of long-term treatment (ie, after the first 3 months), prevention of new episodes of VTE predominates. We use the term extended anticoagulation to refer to anticoagulation that is continued beyond 3 months without a scheduled stop date.

So there is no longer a category distinction between 3 months, 6 months and “indefinite.” Here are the recommendations:


Association with transient risk factor (provoked): 3 months.


No transient risk factor (unprovoked): extended. There is no predefined stop time but upon periodic reassessment the clinician and patient may decide to stop at some point. (If the patient is at high bleeding risk the recommendation is for 3 months).


The recommendation is similar for a recurrent unprovoked DVT.


LMWH is recommended over warfarin for long term treatment of patients with DVT and cancer.


Prevention of PTS: compression stockings for all patients with acute symptomatic DVT, worn for at least 2 years.


PE treatment
For treatment of PE the recommendations are similar to those for DVT with the following differences:


Outpatient treatment is not recommended but early discharge is suggested for patients at low risk.


Who gets thrombolytic therapy? Absent a high bleeding risk the authors suggest systemic thrombolysis for those with hypotension and for selected patients without hypotension who are clinically deemed at high risk for developing hypotension. This latter recommendation is vague. The authors advocate for patient selection in non hypotensive patients based on an overall clinical assessment rather than any particular test or battery of tests.


Concerning catheter interventions and surgical embolectomy for PE the guidelines state:

5.7 Catheter-Based Thrombus Removal for the Initial Treatment of Patients With PE

5.7. In patients with acute PE associated with hypotension and who have (i) contraindications to thrombolysis, (ii) failed thrombolysis, or (iii) shock that is likely to cause death before systemic thrombolysis can take effect (eg, within hours), if appropriate expertise and resources are available, we suggest catheter-assisted thrombus removal over no such intervention (Grade 2C).

5.8 Surgical Embolectomy for the Initial Treatment of Patients With PE

5.8. In patients with acute PE associated with hypotension, we suggest surgical pulmonary embolectomy over no such intervention if they have (i) contraindications to thrombolysis, (ii) failed thrombolysis or catheter-assisted embolectomy, or (iii) shock that is likely to cause death before thrombolysis can take effect (eg, within hours), provided surgical expertise and resources are available (Grade 2C).

The guidelines recommend insertion of an IVC filter for acute PE when a contraindication to anticoagulation exists. (Note that the Thrombosis Interest Group of Canada recommends against this indication for an IVC filter unless an acute proximal DVT is also demonstrated).


Thromboembolic pulmonary hypertension
For chronic thromboembolic pulmonary hypertension: extended anticoagulation, consider surgical thromboendarterectomy in selected cases such as central disease with appropriate expertise available.


Lower limb superficial vein thrombosis
For lower limb superficial vein thrombosis of at least 5 cm in length the guidelines suggest a prophylactic dose of anticoagulant, preferably fondaparinux (2.5 mg daily) for 45 days. That is based on this study which I blogged here. (Contrast the guideline recommendation with the recommendation for full systemic anticoagulation of SVT found in this recent expert review).


Upper extremity DVT Upper extremity DVT
If it is in the axillary vein or more proximal, 3 months systemic anticoagulation whether or not the DVT is associated with an IV catheter. However if the catheter is not removed (and the authors do not recommend removal of the catheter if it is functional and there is ongoing need for it) then anticoagulation continues for the length of time the catheter is in place if over 3 months. The authors do not recommend regional thrombolysis for UEDVT although they acknowledge that in patients with risk factors for PTS patient preferences and values may dictate thrombolysis as a reasonable alternative.


Clots in unusual places
Clots in veins draining the abdominal viscera:  anticoagulation if symptomatic, not if incidentally discovered.


HIT
Alternative form of anticoagulation (e.g lepirudin, argatroban, danaparoid) for HIT with or without thrombosis.


Platelet transfusions for HIT? Only if bleeding or for a procedure with high bleeding risk.


When to transition to warfarin? Only after substantial platelet recovery (150K).


For patients with HIT or a history of same who require PCI:

..we suggest the use of bivalirudin (Grade 2B) or argatroban (Grade 2C) over other nonheparin anticoagulants.

For patients with a remote history of HIT requiring acute anticoagulation: fondaparinux.


Atrial fibrillation
The below recommendations for anticoagulation include patients with paroxysmal AF and patients with atrial flutter, as well as patients being managed on a rhythm control strategy.


Anticoagulation is indicated for patients with a CHADS 2 score of 1 or above, and for those with mitral stenosis. Dabigatran is suggested over warfarin for this purpose for patients who meet the labeling criteria for dabigatran with the exception of patients with coronary artery disease on antiplatelet therapy, in whom warfarin is preferred.

What about cardioversion? In all cases anticoagulate 4 weeks after regardless of CHADS 2 score, and follow the above recommendations re long term. If elective and greater than 48 hrs duration (or unknown duration) anticoagulate 3 weeks prior or do TEE guided CV. If known 48 hours or less, or in cases of emergency cardioversion, then immediately parenterally anticoagulate and cardiovert.

PFO and atrial septal aneurysm
Asymptomatic: no antithrombotic therapy.
Cryptogenic stroke, first episode without DVT: ASA
Cryptogenic stroke, repeat episode on ASA or with DVT: warfarin and consideration for closure therapy.

Intensity of warfarin anticoagulation for patients with mechanical valves
Aortic without additional risks: INR 2-3 All others 2.5-3.5

Low dose aspirin added in all if low bleeding risk.