Research questions
The aim of this systematic review is to evaluate the safety and efficacy of antiplatelet therapy, in addition to standard of care, compared to no antiplatelet therapy, in adult patients with new-onset or relapsing GCA. The proposed systematic review will aim at answering the following questions:
-
In adult patients with new-onset or relapsing GCA, does antiplatelet therapy reduce GCA-related ischemic complications (permanent blindness, stroke, myocardial infarction, and ischemic event related deaths)?
-
In adult patients with new-onset or relapsing GCA, does antiplatelet therapy increase the risk of major bleeding events?
Eligibility criteria
Eligibility criteria for studies based on patient population, intervention or exposure, comparator and methods are presented in Table 1. We will include studies irrespective of the reported outcomes. We will include randomized controlled trials (RTCs), quasi-randomized trials and non-randomized intervention studies. For observational studies, we will include cohort studies and case-control studies. We will exclude cross-sectional, case series and case reports.
Table 1
– Eligibility Criteria for Studies
Population | Inclusion criteria • Adult patients (≥ 18 years), with new-onset or relapsing GCA based on one of the official ACR classification criteria (version 1990 or version 2022). * |
| Exclusion criteria • Patients with systemic vasculitides other than GCA. |
Intervention/ Exposure | Inclusion criteria • Administration of an oral antiplatelet medication in addition to GCA standard of care. • Accepted antiplatelet medications: aspirin (≥ 80mg daily), clopidogrel (75mg daily), ticagrelor (90mg twice a day) or prasugrel (10mg daily). • Accepted timing of antiplatelet initiation: o Within 8 weeks of GCA onset or relapse. o Already administered at the time of GCA onset or relapse. |
| Exclusion criteria • Antiplatelet medication initiated in a patient with inactive GCA. • Use of oral, intravenous, or subcutaneous anticoagulants. |
Comparator | Inclusion criteria • Absence of adjunctive antiplatelet therapy. |
| Exclusion criteria • Use of oral, intravenous, or subcutaneous anticoagulants. |
Method/ Design | Inclusion criteria • RTCs, quasi-randomized trials, non-randomized intervention studies, cohort studies and case-control studies. • Study duration of at least 6 months. |
| Exclusion criteria • Cross-sectional, case series and case reports. |
GCA: giant cell arteritis; ACR: American College of Rheumatology; RCT: randomised controlled trial. |
* New-onset GCA is defined as a diagnosis of active GCA in a participant with no previous history of GCA. A relapsing GCA is defined as a diagnosis of active GCA in a patient with a history of GCA in remission (asymptomatic). |
Study duration of at least 6 months will be required for study inclusion. End points will be evaluated at 6 and 12 months.
We will include trials irrespective of the language of publication or format in which they were reported. If we identify studies with unpublished data, they will be considered for inclusion. There will be no restriction by type of setting (academic hospital, community hospital, inpatient or outpatient).
Outcomes
The primary composite efficacy outcome is the incidence (proportion) of GCA-related ischemic events occurring during the follow-up period, which include ischemic strokes, permanent blindness, myocardial infarction, or ischemic event-related deaths. The definition of ischemic events is based on the official American Heart Association (AHA) and American Stroke association (ASA) consensus. Ischemic stroke is defined as an episode of neurological dysfunction caused by focal cerebral, spinal, or retinal infarction. Permanent blindness is a permanent loss of sight, whether it be a full or partial loss. Myocardial infarction is defined as an episode of myocardial injury (elevated cardiac troponin values at least above the 99th percentile upper reference limit) with clinical evidence of at least one acute myocardial ischemia manifestation (symptoms of cardiac ischemia, new ischemic ECG changes, new pathological Q waves, compatible cardiac imaging, or coronary thrombus on angiography).
The main secondary efficacy outcome is the incidence (proportion) of ischemic strokes, permanent blindness, myocardial infarction, and death, measured separately and occurring during the follow-up period.
The main safety outcome is the incidence (proportion) of major bleeding events. A major bleeding event is defined based on the classification of the International Society of Thrombosis and Hemostasis as either: fatal bleeding, and/or symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intra-articular or pericardial, or intramuscular with compartment syndrome, and/or bleeding causing a fall in hemoglobin levels of 20 g/L or more, and/or leading to a transfusion.
The main secondary safety outcome is the incidence (proportion) of death due to a bleeding event.
Information sources
Published studies and unpublished gray literature will be searched. We will search the following electronic bibliographic databases: MEDLINE (Ovid interface, January 1946 – onwards), Cochrane Central Register of Controlled Trials in the Cochrane Library (CENTRAL) and EMBASE (Ovid interface, January 1947- onwards). The following study registries will be searched: metaRegister of controlled trials (mRCT) and ClinicalTrials.gov.
Online conference papers, abstracts, and presentations from the American College of rheumatology annual meeting (ACR, from 2012 – onwards) and European League against Rheumatism annual meeting (EULAR, from 2001 – onwards) will be searched.
References will be screened to make a list of experts in the field. A citation index search of these experts will be performed on the ‘Web of Science’ platform and ‘researchgate.net’. Finally, a manual bibliography search of retrieved records will be performed to find additional references.
We will search for retraction or errata statements that were published for every study we include.
Search strategy
Literature search strategies was developed using medical subject headings (MeSH) and text words related to GCA and antiplatelet therapy. There will be no language restriction for the search. A publication date filter from August 1st, 1990 - onwards will be used because official classification criteria for GCA were released in August 1990. This will allow correct identification of the population of interest.
The search strategy was elaborated by the corresponding author and was peer-reviewed by all authors. A transcript of the MEDLINE search strategy is provided (Fig. 1). Search strategies for other information sources are provided in the data supplement (figure S1 to S5). We will provide the actual date when each search was performed during the review stage.
Study records management, selection, and collection
Covidence (covidence.org) will be used to upload literature search results, manage study records, select studies and for data extraction. Two co-authors (OT, YB) will work independently for the initial screening (title and abstract), selection of studies and data extraction. The two reviewers will have a practice-run with 10 records as a calibration exercise before beginning the official review process.
Duplicate records will be counted and removed using Covidence. Multiple reports of the same study will be identified by juxtaposing author names, location, setting and date of the study. Multiple reports of the same study will not be discarded; they will be collated under one study identification. However, one record will be selected as the source (main report) of the study with a justification provided.
Any disagreement between the two reviewers will be resolved through discussion. A third reviewer (CR) will be consulted if the disagreement persists. If required, additional information from the study authors will be requested to resolve remaining questions on eligibility.
Data items
Extracted data will include identification information (study id, record id), reason for excluding studies, characteristics of the study (author, year, country, design, duration, funding, conflict of interest), characteristics of participants (age, sex, ethnicity, disease subtype, baseline characteristics), details of the intervention (timing of initiation, type of antiplatelet medication, dose, other GCA therapy and immunosuppression), details of the comparator, outcomes (definitions, method of aggregation, measures of association, timing), and results (number of participants, exclusion, losses at follow-up, summary results, subgroup results). Furthermore, key conclusions, comments and references will be collected.
Missing data will be recorded as such. Study authors will be contacted to retrieve missing data or resolve any uncertainties. Communications with study authors will be performed by email. A maximum of 3 attempts to reach the authors will be made in case of no response, with each attempt every 14 days.
Risk of bias of individual studies
For randomized studies, we will use version 2 of the Cochrane Collaboration tool for assessing the risk of bias (RoB 2).(22) For non-randomized intervention studies, the Risk Of Bias In Non-randomized Studies - of Interventions (ROBINS-I) tool will be used.(23) For observational studies, the Risk Of Bias In Non-randomized Studies - of Exposure (ROBINS-E) tool will be used.(24)
The risk of bias will be assessed independently by two reviewers (OT, YB). Disagreements will be resolved by discussion, and if required, with arbitration from a third reviewer (JPM). Using the software RevMan web, we will summarise our findings and present them in a risk of bias table and figure.
Data synthesis
If included studies are sufficiently homogeneous in terms of participants (similar proportions of new-onset or relapsing GCA), interventions (antiplatelet type), comparator, and study design (duration of follow-up), we will perform meta-analyses using a random-effects model. Our data will be binary: proportions of GCA-related ischemic complications, major bleeding event and deaths. We will use a Mantel-Haenszel method for quantitative synthesis, with odds ratio (OR) as a measure of association with 95% confidence intervals.
A sensitivity analysis will be performed to evaluate the impact of studies with significant missing data (when there is ≥ 10% missing data for any outcome).
Statistical heterogeneity will be tested using Chi2 test (significance level of 0.1) and I2 statistic. If there is a high level of statistical heterogeneity (p < 0.1 or I2 > 50%), we will analyze clinical heterogeneity by documenting the variability in participants, interventions, and outcomes in the included trials. We will also carefully analyze and compare study designs and settings to assess methodological heterogeneity.
Furthermore, we will perform the following subgroup analyses to better understand the source of heterogeneity: 1) subgroup based on disease subtype (new onset vs. relapsing), 2) based on the timing of antiplatelet medication (prior to GCA vs. at GCA onset or relapse), 3) based on GCA therapy received (glucocorticoids alone vs. glucocorticoids with immunosuppression).
We will also perform the following sensitivity analyses: 1) exclusion of non-randomized studies, 2) exclusion of studies with a high risk of bias.
If quantitative analysis is not appropriate, a systematic narrative synthesis will be provided.
Meta-biases assessment
Publication bias will be evaluated with a funnel plot if at least 10 studies are included in the meta-analysis. We will use Egger’s test to assess potential publication bias via funnel plot asymmetry. For each included study, outcome reporting bias will be investigated by comparing reported outcomes against planned outcome measures “a priori” in the study protocol or trial registry.
The risk of bias due to selective outcome reporting will be considered low if 1) the study protocol was published before the availability of study results, and 2) every “a priori” outcome in the protocol is reported in the study record (or if justification was provided for not reporting an outcome).
If no study protocol is found, we will use the Outcome Reporting Bias in Trials (ORBIT) classification system to evaluate the risk of selective outcome reporting.(25)
A sensitivity analysis to assess the impact of selective outcome reporting will be conducted if at least one study presents a high risk of bias due to selective outcome reporting.
Confidence in cumulative evidence
Two reviewers (JPM, CR) will independently evaluate the quality of evidence for all outcomes using the Grading of Recommendations, Assessment, Development and Evaluation working group methodology (GRADE).(26) Conclusions on GRADE for each outcome will be displayed in a ‘summary of findings’ table. An overall GRADE will be given to the body of all outcomes.