Antiplatelet and anticoagulant agents for preventing recurrence of peripheral vascular thrombosis in patients with Antiphospholipid syndrome
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the efficacy and safety of antiplatelet and anticoagulant agents for preventing the recurrence of any peripheral vascular thrombosis (arterial or venous, or both) in patients with Antiphospholipid syndrome.
Background
Description of the condition
Antiphospholipid syndrome (APS) or 'Hughes syndrome' is a prothrombotic disorder characterized by vascular thrombosis (venous or arterial, or both) or pregnancy complications, or both, in association with persistently positive antiphospholipid auto‐antibodies (aPLs), namely lupus anticoagulant (LA), anticardiolipin antibodies (aCL) or anti‐β2 glycoprotein I (β2GPI) antibodies (Cervera 2002; Miyakis 2006; Asherson 2009). These auto‐antibodies bind to plasma proteins found on negatively‐charged phospholipids, such as phosphatidylserine or cardiolipin, which contribute to the formation of thrombi and subsequently thromboembolism.
The syndrome can be categorized as primary APS (PAPS), where there is no association with other known autoimmune diseases (Asherson 1989), or as APS associated to other diseases — particularly autoimmune diseases such as systemic lupus erythematosus (SLE) or other less common disorders, such as systemic sclerosis, rheumatoid arthritis or Behçet's syndrome (Alarcón‐Segovia 1989; Greaves 1999). Catastrophic APS (CAPS) is another type of APS, associated with high mortality, where blood clots appear in multiple small vascular beds that lead to multiple organ failure (Cervera 2002; Cervera 2009).
Thrombosis (both venous and arterial) is one of the two major clinical features in both the Sapporo (Wilson 1999) and Sydney (Miyakis 2006) classification criterion of APS. Initially thrombosis was documented as the most frequent clinical event (30%) in APS patients (Lechner 1985). After 17 years of initial assessment, Cervera 2002 established deep venous thrombosis (DVT) as the most commonly observed clinical manifestation followed by thrombocytopaenia, livedo reticularis, stroke, superficial thrombophlebitis, pulmonary embolism (PE), fetal loss and transient ischemic attack (TIA). Again in 2005, thrombosis (both venous and arterial), was the most common clinical manifestation related to highest morbidity (Santamaria 2005). Venous thrombosis accounts for approximately two‐thirds of all the thromboembolic events, represented mainly by DVT of the legs and PE (Barbui 1993). During the course of the disease, approximately 55% of patients with APS suffer from DVT of the extremities and 40% suffer PE (Asherson 2002). In the Euro‐Phospholipid project, DVT of the lower extremities was the most common clinical manifestation of APS. However, any tissue or organ vascular bed can be affected. During the 10‐year period study on 1000 patients across 13 European countries (Euro‐Phospholipid project), thrombosis was the most frequent manifestation (28.1%) with 5.3% stroke, 4.7% TIA, 4.3% DVT, 3.5% PE, 8.1% livedo reticularis and 3.1% skin ulcer (Cervera 2014). The incidence of first thrombosis was as high as 5.3% per year when the patient was detected with all three aPLs (triple positivity) (Pengo 2011). Approximately 9.5% APS patients with DVT exhibited aPLs (Gómez‐Puerta 2014).The recurrence of thrombosis in APS patients is a common phenomenon with high occurrence rate. A study with APS patients (N = 147) revealed 69% recurrence (undergoing treatment with warfarin or low dose aspirin, or both) of thrombosis (52% arterial, 40% venous and both 8%), which took place within an average duration of 12 months from the first occurrence (Khamashta 1995). Another study (N = 114) showed recurrent thrombosis occurred in 10.7% of participants assigned to receive high intensity amounts of warfarin and in 3.4% of participants who were assigned to take moderate intensity amounts of warfarin (Crowther 2003).
Even though the prevalence of positive aPLs in the general population is as high as 5%, only a few individuals develop APS. The prevalence of APS is 40 to 50 cases per 100,000 people per year. The incidence is approximately five new cases per 100,000 people per year (Cervera 2002). The male:female ratio was estimated to be 1:3.5 for PAPS and 1:7 for APS associated to other diseases (Cervera 2002). However, another study (N = 121) estimated the male:female ratio as almost 1:1 (Avčin 2008). The total mortality rate during a 10‐year observational period with APS patients was 9.3%. However, in the case of CAPS it was up to 55.6%. The most common cause of death was bacterial infection (21.5%), followed by myocardial infarction (13.9%), stroke (11.8%), hemorrhage (10.7%), malignancy (13.9%), CAPS (5.4%) and PE (5.4%) (Cervera 2014).
A strong association between aPLs and thrombosis has already been established; however the pathogenic role of aPLs in the development of thrombosis is yet to be clarified. According to the latest classification criteria (Miyakis 2006), the presence of either of the three aPLs is considered a positive laboratory feature for confirming APS. However, there is a tendency of higher occurrences of thrombotic events when multiple aPLs are present (Forastiero 2014). So far, several discrete pathways have been described as the potential pathogenic mechanisms of thrombosis in APS. In thrombi generation, different studies have indicated the pivotal roles of APS associated autoantibodies which direct against a number of plasma proteins and proteins expressed on, or bound to, the surface of vascular endothelial cells or platelets (Robertson 2006). The aPLs have crucial effects upon platelet activation and pathways of coagulation including procoagulant actions of these antibodies upon protein C, annexin V, serum proteases, toll‐like receptors, tissue factor and via impaired fibrinolysis (Forastiero 2008; Kinev 2008; Rand 2008; Raschi 2008; Urbanus 2008; Bu 2009; Chen 2010). Changes in the behavior of cells and the secretion of various immune‐associated molecules occur due to the involvement of aPLs in clinically‐important normal procoagulant and anticoagulant reactions. This might be the basis of possible mechanisms by which aPLs can lead to the development of thrombotic events in APS patients (Espinosa 2003; de Groot 2005; Giannakopoulos 2013).
Description of the intervention
The optimal treatment for APS is still unclear. Currently the regimens used for the treatment of APS include: (1) warfarin, (2) antiplatelet agents (aspirin or clopidogrel) and (3) low molecular weight heparin (LMWH) (Hughes 2010), which is the same as the treatment offered to the general population presenting with thrombotic events (Petri 2001). Among the newly licensed (European Medical Agency) and approved (Food and Drug Administration) oral anticoagulants, rivaroxaban, apixaban, edoxaban and dabigatran are well‐known (Erkan 2014). These agents, unlike warfarin, are fixed dose with predictable anticoagulant effects and there is no need to monitor anticoagulant intensities routinely. These do not interact with dietary constituents or alcohol, and have a few reported drug interactions that affect anticoagulant intensity (Erkan 2014).
Warfarin
Warfarin is one of the most widely used anticoagulants internationally and is typically used to prevent thrombosis and thromboembolism. After oral administration, it is rapidly absorbed from the gastrointestinal (GI) tract and reaches maximal blood concentrations in healthy adults after 90 minutes (O'Reilly 1976; Breckenridge 1978). It is a racemic mixture of two active stereoisomers (R and S) with a half‐life of 36 to 42 hours. It circulates by binding with plasma proteins (mainly albumin) and accumulates in the liver where metabolic transformation takes place mainly by different cytochrome P450 enzymes via different pathways (Kaminsky 1997; Wessler 2013). A recommended international normalized ratio (INR) intensity is between 2.0 and 3.0. However, patients with APS may require a higher targeted INR than 2.0 to 3.0 (Hirsh 2001). A retrospective study with APS patients showed higher efficacy of anticoagulation to a target INR greater than 3.0 in preventing recurrent thrombosis compared with other regimes including standard intensity (i.e. target INR 2.0 to 3.0) anticoagulation or antiplatelet therapy (Khamashta 1995). Recently, long‐term anticoagulation has also been recommended in APS with thrombosis to prevent recurrences (Erkan 2014). Warfarin may cause severe bleeding that can be life‐threatening and even can cause death (Wysowski 2007).
Heparin
Heparin is another common (injectable) anticoagulant which is widely used. The bioavailability of LMWH is 100% (Boneu 1990), and the half‐life is four to five hours (Weitz 2004). In the treatment of APS the dose of LMWH (enoxaparin) is usually 1 mg/kg/day (Di Prima 2011; Danowski 2013). In patients with thrombosis who receive high doses, heparin‐induced bleeding, thrombocytopaenia and increased transaminases are the most common adverse effects (Hirsh 1983; Monreal 1989).
Rivaroxaban
Rivaroxaban is absorbed rapidly with no effect of food on absorption or pharmacokinetic parameters. The maximum plasma concentrations peak is observed at one to four hours (Kubitza 2005). A 10 mg tablet has a bioavailability of approximately 80% (Eriksson 2009). In young adults the plasma elimination half‐life is five to nine hours and 11 to 13 hours in older people (due to the age‐related renal dysfunctions) (Kubitza 2005; Mueck 2008). It is metabolized via CYP3A4, CYP2J2 and CYP‐independent mechanisms (Weinz 2009). Even though serious and fatal bleeding are quite uncommon, higher rates of bleeding in the GI tract are commonly seen (Brown 2015). A recent case series (n = 8) revealed that there was no recurrence of thrombotic events in any of the APS patients during treatment (nine to 36 months) with rivaroxaban (20 mg/day) (Betancur 2016). However, according to another case series (n = 8), a decreased protective effect of rivaroxaban was observed in APS patients with triple positivity (LA, aCL and β2GPI). Therefore, until the results of an ongoing trial (randomized controlled phase II/III clinical trial) on rivaroxaban are available (Cohen 2015), the efficacy of rivaroxaban remains inconclusive.
Apixaban
Apixaban is an anticoagulant drug with rapid oral absorption. Maximum plasma concentrations peak is achieved by one to three hours after administration (Frost 2013). Apixaban has a bioavailability of approximately 66% (Frost 2008). The drug has an elimination half‐life of eight to 15 hours in healthy young adults (Frost 2007). Apixaban is metabolized mainly via CYP3A4 and also to a lesser extent by CYP1A2, 2C8, 2C9, 2C19 and 2J2 (Wang 2011). For the prevention of venous thromboembolism (VTE), a dose of 2.5 mg twice daily is suggested (Lassen 2009). Apixaban has a lower risk of bleeding when compared with warfarin (Hylek 2014). According to a recent report (Betancur 2016), recurrence of thrombosis was not observed in a patient with APS (suffering from venous thrombosis in presence of LA and aCL (both IgG and IgM)) treated with apixaban (5 mg three times a day). However, until the results of an ongoing prospective pilot study (Woller 2016) (randomized open‐label blinded) in APS patients are available, the efficacy of apixaban in prevention of secondary thrombotic events remains unclear.
Edoxaban
Edoxaban is another new anticoagulant drug that shows rapid oral absorption, with the maximum plasma concentrations achieved at one to two hours after administration. It has a bioavailability of approximately 60% with a mean half‐life of six to 10 hours in healthy young adults. Edoxaban is metabolized via CYP3A4 with an oral bioavailability of approximately 60%. For preventing VTE, 30 mg edoxaban is given once daily (Bounameaux 2014). The most prevalent adverse effect observed was bleeding (Minor 2015).
Dabigatran
Dabigatran is an oral anticoagulant drug with low bioavailability (approximately 6% to 7%) (Blech 2008). It has a half‐life of 12 to 14 hours and the maximum plasma peak is achieved within two hours of oral administration (Stangier 2008). Dabigatran is not metabolized by the CYP enzymes. However, it is a substrate for the P‐glycoprotein transporter and mainly excreted by the kidneys (Brighton 2010). The usual recommended oral dose of dabigatran is 110 mg or 150 mg, twice daily and the common adverse effect is bleeding (Wallentin 2010). Some studies indicate dabigatran is a successful new oral anticoagulant drug in the prevention of recurrence of thrombotic events in patients with APS (Noel 2015; Reshetnyak 2015), while others show dabigatran is unsuccessful in preventing recurrent thrombosis in APS patients (Schaefer 2014) or report adverse neurological abnormalities such as acute vision changes, transient weakness, presyncopal events and memory loss (Win 2014).
Aspirin
Aspirin (acetyl salicylic acid) is an antiplatelet agent which is commonly used to prevent thrombosis. When it is administered orally, it reaches its peak plasma level approximately 30 to 40 minutes after ingestion and is rapidly absorbed in the stomach and upper GI tract (Benedek 1995). Measurable inhibition of platelet function achieved within 60 minutes after aspirin ingestion (Jimenez 1992; Patrono 2001). Enteric‐coating of aspirin significantly delays the time (three to four hours) to reach its peak plasma effect (Patrono 2001; Hirsh 2004). The plasma half‐life of aspirin is only 20 minutes. The oral bioavailability of regular aspirin tablets is approximately 40% to 50% over a wide range of doses (McAdam 1999). Aspirin acts on platelet function by permanently inactivating a key platelet enzyme, cyclooxygenase (COX) (Roth 1975). However, this effect can be reversed by the generation of new platelets (Burch 1978). After a single dose of aspirin, platelet COX activity recovers by approximately 10% per day due to platelet turnover (Burch 1978), therefore, once‐daily dosage is required. Results from randomized controlled trials (RCTs) reported that aspirin is an effective antithrombotic agent when the dosage is between 50 and 100 mg/day, and the suggestible effective dosage is as low as 30 mg/day (Patrono 2001). The major adverse effect of aspirin administration is an increased risk of bleeding complications, most commonly GI tract bleeding (Patrono 2001). Administration of nonselective reversible COX inhibitors (i.e. ibuprofen and naproxen) with aspirin results in competition for a common docking site (arginine 120) within the COX channel. This may impair the efficacy of aspirin (Catella‐Lawson 2001; Capone 2005). Clinicians usually recommend low‐dose aspirin for the primary prevention of thrombosis in asymptomatic patients with moderate to high levels of aPLs (Farmer‐Boatwright 2009). Recently a patient with APS suffering from livedoid vasculitis was successfully treated with aspirin (100 mg/day) in combination with warfarin (2 mg/day) to prevent the recurrence of thrombosis (So 2015).
Clopidogrel
Prevention of recurrent thrombosis in patients with APS is also managed by an antiplatelet agent known as clopidogrel (Nalli 2014). Clopidogrel is a prodrug (a precursor chemical compound of a drug) which is rapidly absorbed in the intestine and is activated in the liver by cytochrome P450 enzymes (CYP1A2, CYP2B6, CYP2C9, CYP2C19 and CYP3A4/5) (Liu 2012). It is undetectable in human plasma, with a half‐life of approximately six hours after a single oral dose of 75 mg (Sangkuhl 2010). Bleeding is one of the most common adverse effects of clopidogrel (Eikelboom 2006). When clopidogrel was used in combination with hydroxychloroquine, recurrence of thrombotic events and bleeding episodes were not observed in a patient with APS (Park 2013).
How the intervention might work
Warfarin
Vitamin K is an essential co‐factor for normal blood coagulation in the postribosomal synthesis of biologically‐active forms of the calcium‐dependent clotting factors II (prothrombin), VII, IX and X, as well as the regulatory factors protein C, protein S and protein Z (Freedman 1992; Ansell 2008). The vitamin K‐dependent clotting factors II, VII, IX and X contain a series of glutamate residues at their amino terminus. For proper functioning of the clotting factors, during posttranslational modification (in the liver), the glutamate residues are carboxylated and converted to γ‐carboxyglutamate (Gla) residues in the presence of O2, CO2 and carboxylase enzyme. The carboxylation process is associated with the vitamin K cycle where oxidation of vitamin K produces vitamin K epoxide (vitamin KO). Vitamin KO is converted back to vitamin K by the enzyme vitamin KO reductase (VKOR). Warfarin interferes in the action of VKOR so that vitamin KO cannot be recycled back to vitamin, thus reducing the availability of vitamin K and the production of blood clotting factors. Due to this disruption, blood clotting and the risk of thrombi generation can be reduced (Hirsh 2001; Jennings 2008).
Heparin
Under normal circumstances, antithrombin III (ATIII) inactivates coagulation enzymes thrombin (factor IIa) and factor Xa. This process occurs at a slow rate. When heparin is administered, it reversibly binds to ATIII and accelerates the rate of coagulation enzymes inactivation by ATIII. The rate of inactivation of these enzymes by ATIII can increase by up to 1000‐fold due to the binding of heparin. Thus the mechanism of action of heparin is ATIII‐dependent. It also prevents the progression of existing clots by inhibiting further clotting (Björk 1982; Jackson 1990).
Rivaroxaban
Factor Xa (FXa) is an essential blood coagulation factor that is responsible for the initiation of the coagulation cascade. FXa activates prothrombin (factor II) to thrombin (factor IIa). Thrombin (a serine protease) then converts fibrinogen to fibrin in the coagulation cascade to activate platelets. Stabilization of the platelet aggregation by fibrin mesh ultimately leads to clot formation. Rivaroxaban is an anticoagulant which directly and irreversibly binds to factor Xa. Thereafter, it effectively blocks the amplification of the thrombin generation and prevents the formation of thrombus (Samama 2011; Graff 2013).
Apixaban
Apixaban exerts no direct effect on platelet aggregation. It acts by inhibiting coagulation through direct, selective and reversible inhibition of free and clot‐bound FXa. It has moderate selectivity for clot‐bound over free FXa. Thus apixaban indirectly decreases platelet formation induced by thrombin (Frost 2013).
Edoxaban
Edoxaban is also a direct inhibitor of FXa (acting as an activator of prothrombinase in converting prothrombin to thrombin) that rapidly and selectively inhibits both free and cell‐bound FXa in a concentration‐dependent manner. Eventually thrombin cannot be formed because of the inhibited FXa, and clot formation is prevented following the same mechanism as rivaroxaban and apixaban (Perzborn 2010).
Dabigatran
Dabigatran etexilate is an inactive pro‐drug that is converted to dabigatran (the active form) by esterase‐catalysed hydrolysis in the plasma and liver. It is a rapid‐acting competitive, selective, reversible and direct inhibitor of both free and clot‐bound thrombin (factor IIa) which acts by binding to the active site of the thrombin molecule. Inhibition of thrombin consequently prevents thrombus development by inhibiting fibrin formation in the coagulation cascade. Thus dabigatran inhibits thrombin‐induced platelet aggregation and the formation of blood clots (Galanis 2011; Ferrandis 2013).
Aspirin
Thromboxane A2 (TXA2) is a type of thromboxane which is produced by activated platelets. It stimulates activation of new platelets and increases platelet aggregation. COX is the key enzyme that catalyses the production of TXA2 from arachidonic acid in platelets. When aspirin is taken as an antiplatelet agent, it directly and irreversibly inactivates the COX enzyme through acetylation so that COX further cannot activate TXA2. As a result, platelet aggregation is inhibited, resulting in less blood clot or thrombi generation (Schrör 1996; Awtry 2000).
Clopidogrel
Clopidogel specifically and irreversibly inhibits the P2Y12 subtype of ADP receptor on the platelet cell membrane, and thus obstructs the binding of ADP to its receptor, blocks ADP‐mediated activation of the glycoprotein GPIIb/IIIa complex (major fibrinogen receptor by which fibrinogen binds to platelets and causes platelet aggregation by cross‐linking) and consequently hinders the platelets activation and aggregation (Vane 2003; Jennings 2008). Thus it reduces the probability of clot formation as well as the chance of thrombosis.
Why it is important to do this review
Several treatment guidelines have recommended warfarin and aspirin as thrombotic prophylaxis in preventing recurrent thrombosis in patients with APS (Keeling 2012; Nalli 2014). However, there is still uncertainty in choosing the appropriate intervention to avoid recurrence of peripheral vascular thrombosis in APS. There is a Cochrane systematic review on aspirin and/or heparin intervention on recurrent miscarriages of APS patients (de Jong 2014). However, the efficacy of antiplatelet and anticoagulant regimens to prevent recurrent peripheral vascular thrombosis remains obscure. A systematic review based on RCTs is required to evaluate the comparative efficacy between antiplatelet and anticoagulant regiments for preventing recurrent thrombosis in patients with APS.
Objectives
To assess the efficacy and safety of antiplatelet and anticoagulant agents for preventing the recurrence of any peripheral vascular thrombosis (arterial or venous, or both) in patients with Antiphospholipid syndrome.
Methods
Criteria for considering studies for this review
Types of studies
We will include all RCTs or quasi‐RCTs. We will include trials that use quasi‐randomized methods, such as alternation, if there is sufficient evidence that the treatment and control groups were similar at baseline. We will not consider cross‐over trials for inclusion.
Types of participants
We will include people with APS (both PAPS and APS associated to other diseases) following the Sapporo (Wilson 1999) or the Sydney criteria (Miyakis 2006).
Types of interventions
-
Interventions: anticoagulant and antiplatelet agents.
-
Comparators: placebo, other anticoagulant or antiplatelet agents and comparisons of the same agents with different dosages, intensities and duration.
Types of outcome measures
Primary outcomes
-
Incidence of recurrent peripheral vascular thrombosis. Diagnosis: thrombosis (arterial or venous) confirmed by objective validated criteria (i.e. unequivocal findings of appropriate imaging studies or histopathology) as described by Sapporo (Wilson 1999) or Sydney (Miyakis 2006) classification criteria.
-
Bleeding (minor and major). Diagnosis: bleeding diagnosis based on the definitions of major and minor bleeding (Schulman 2005).
-
Death from major peripheral vascular thrombosis.
Secondary outcomes
-
Quality of life, measured either qualitatively or quantitatively (Alba 2016).
-
Activated Partial Thromboplastin Time (APTT), measured quantitatively using standardized methods (Arnout 1999).
Search methods for identification of studies
Electronic searches
The Cochrane Vascular Information Specialist (CIS) will search the Specialised Register and the Cochrane Register of Studies (CRS) (http://www.metaxis.com/CRSWeb/Index.asp). See Appendix 1 for details of the search strategy which will be used to search the CRS. The Specialised Register is maintained by the CIS and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL, AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used, are described in the Specialised Register section of the Cochrane Vascular module in the Cochrane Library (www.cochranelibrary.com).
The CIS will search the following trial databases for details of ongoing and unpublished studies using the term 'antiphospholipid syndrome'.
-
The World Health Organization International Clinical Trials Registry (http://apps.who.int/trialsearch/).
-
ClinicalTrials.gov (http://clinicaltrials.gov/).
-
The ISRCTN Register (http://www.isrctn.com/).
Searching other resources
We will check the citations of relevant trials retrieved by electronic searches for additional relevant studies. We will contact the authors of relevant articles or ongoing trials by email to request data or papers to identify any unpublished RCTs.
Data collection and analysis
Selection of studies
Two review authors (MAI and FA) will independently evaluate the identified abstracts and full‐text articles from the databases and other sources based on the criteria for inclusion of relevant studies in the Cochrane review. If any disagreements arise on the eligibility of a study for inclusion, we will make a decision by discussion with a third review author (THS). We will describe the reasons for exclusion of studies in the review.
Data extraction and management
One review author (FA) will independently perform data extraction and another review author (MAI) will verify all the extracted data for accuracy and consistency. We will resolve any disagreements by discussion between the review authors.
Assessment of risk of bias in included studies
Three review authors (THS, AH and SA) will independently assess the risk of bias in included studies using the Cochrane 'Risk of bias' assessment tool, as described in Section 8.5 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will resolve disagreements by discussion with the review authors.
Measures of treatment effect
We will use the risk ratio (RR) with 95% confidence intervals (CIs) for dichotomous data and the mean difference (MD) with 95% CIs for continuous data in case of identical scales. However, when there is a difference of the scales without changing the measured outcome, we will use the standardized mean difference (SMD) with 95% CIs.
Unit of analysis issues
The unit of analysis will be the participating individuals in the randomized trials. The outcomes of death and bleeding (both major and minor) are dichotomous and we will measure the incidence rate as per 100 patient per year outcome.
Dealing with missing data
We will perform intention‐to‐treat analysis where needed. We will only analyse the available data. However, if we notice data are missing, we will contact the authors of the relevant study to request the missing data. If unsuccessful, we will perform sensitivity analysis to investigate any effect on the meta‐analysis.
Assessment of heterogeneity
We will assess heterogeneity using the Chi² test and the I² statistic as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will define moderate and substantial heterogeneity as an I² statistic value of greater than 30% and less than 50%, and an I² statistic value greater than 50%, respectively.
Assessment of reporting biases
We will use funnel plots to assess publication and reporting bias when we include more than 10 studies in the meta‐analysis (Higgins 2011).
Data synthesis
We will use Cochrane's statistical software, Review Manager (RevMan) (RevMan 2014), to analyse data based on the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will consider a fixed‐effect model where we find no substantial heterogeneity (I² statistic is less than 50%). We will use a random‐effects model if we find substantial heterogeneity (I² statistic is greater than 50%).
Subgroup analysis and investigation of heterogeneity
If we identify substantial heterogeneity (I² statistic is greater than 50%) between studies, we will perform subgroup analyses on the following groups: age; sex; ethnicity; PAPS or APS associated to other diseases; presence of single or multiple aPLs; duration and response to anticoagulant and antiplatelet drugs; and arterial and venous thrombosis. Additionally, we intend to investigate different effects of anticoagulants and antiplatelets on the stated subgroups.
Sensitivity analysis
We will perform sensitivity analysis to verify the robustness of results from studies where included studies use quasi‐randomization methods or if there are variations among one or more inclusion criteria of the included studies. We will also perform sensitivity analysis by excluding studies that are at high risk of bias.
'Summary of findings' table
We will construct a 'Summary of findings' table using the GRADEpro Guideline Development Tool (www.gradepro.org) to present the main findings of the review. People with APS will constitute the studied population. We will make comparisons of antiplatelet and anticoagulant agents with placebo, other anticoagulant or antiplatelet agents, or the same agent with different dosages, intensities and duration. We will use weighted mean numbers of events in the control group of the included studies to calculate assumed control intervention risks. We will present typical risks for participants that receive the control intervention (placebo, other anticoagulant or antiplatelet agents) of the number of people that experience the event per 1000 people. We will assess the quality of the evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (Atkins 2004). We will include the main outcomes listed in the 'Types of outcome measures' section that we consider essential for decision‐making in the 'Summary of findings' table.
