Description of the condition
Cardiovascular disease is the leading cause of morbidity and mortality among people at all stages of chronic kidney disease (CKD) (Casas 2005; Keith 2004; Mann 2001; Norris 2006; Sarnak 2003; Weiner 2004a; Weiner 2004b) including kidney transplant recipients (Aakhus 1999; ANZDATA 2009; Kasiske 2000; Ojo 2000; USRDS 2010). Compared with the general population, the risk of cardiovascular disease is increased two-fold in people with the early stages of CKD (Go 2004) and 30- to 50-fold in people who need dialysis (de Jager 2009; Fort 2005) in whom it accounts for half of all deaths (Collins 2003). Population representative surveys in Australia (AusDiab 2003) and the United States (NHANES 2010) have shown that CKD (defined as proteinuria or reduction of glomerular filtration rate (GFR) below 60 mL/min/1.73 m²) affects approximately 16% of the adult population. With the increasing prevalence of some of the known risk factors for CKD, including hypertension, obesity and diabetes (Fields 2004; Koren-Morag 2006; Mokdad 2003), the burden of CKD and its complications are projected to increase and to contribute significantly to global healthcare expenditure.
How the intervention might work
Antiplatelet agents prevent arterial occlusion from thrombus via direct prevention of platelet aggregation. Current available data suggest antiplatelet agents might be beneficial in patients with CKD for primary (ATT 2002; HOT Study 2010; Ruilope 2001) and secondary (Berger 2003; McCullough 2002) prevention of cardiovascular events. Antiplatelet agents may have beneficial effects on the kidney, possibly reducing proteinuria and protecting kidney function in people with glomerulonephritis (Taji 2006; Zäuner 1994), and improving graft function in kidney transplant recipients (Bonomini 1986; Frascà 1986). However, some have reported that the efficacy of antiplatelet therapy in CKD might be lower than for other high cardiovascular risk populations (Best 2008). Despite this, the Kidney Disease Outcomes Quality Initiative guideline program (KDOQI) has supported the use of aspirin for primary prevention of cardiovascular disease in CKD. Antiplatelet agents appear to have a modest effect on the preservation of arteriovenous fistula patency (Dember 2008). Their use for fistula preservation and as part of a multifactorial intervention strategy for patients with CKD is advocated by guideline groups (CARI 2000; Renal Association 2010).
Why it is important to do this review
To date, there has been no formal meta-analysis of the benefits and harms of antiplatelet agents in patients with CKD. In contrast to the general population, people with CKD have a different profile of causes for major cardiovascular events, including a greater preponderance for arrhythmia and congestive heart failure (Amann 2003; Curtis 2005; Dikow 2005; Foley 1995; Remppis 2008), altered pharmacokinetics (Mosenkis 2004; Scheen 2008) and impaired haemostasis (Kaw 2006; Remuzzi 1988; Wattanakit 2008; Zwaginga 1991). Compared with people who do not have CKD, these factors might expose the CKD population to a different spectrum of risk and benefit from antiplatelet therapy.
To evaluate the benefits and harms of antiplatelet therapy in people with any form of CKD, including those with CKD not receiving renal replacement therapy, patients receiving any form of dialysis, and kidney transplant recipients.
Criteria for considering studies for this review
Types of studies
All randomised controlled trials (RCTs) and quasi-RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) of antiplatelet agents in people with CKD were included.
Types of participants
Participants with CKD, including those who needed renal replacement therapy (dialysis), had a functioning kidney transplant, or whose kidney function was impaired (defined as a reduced GFR < 60 mL/min/1.73 m²), the presence of other markers of kidney damage such as proteinuria (KDOQI stages 1 to 5), or an elevated serum creatinine (SCr) level (SCr > 120 μmol/L). Data from subgroups of participants with CKD within studies with broader inclusion criteria (e.g. people from the general population, people with diabetes, people with cardiovascular disease) were also included.
Types of interventions
Interventions included any antiplatelet agent. Agents could be administered at any dose or route of administration, and compared with placebo, no treatment, different dose of the same or different antiplatelet agents, different administration regimens of the same or different agent, or different combinations of antiplatelet agents. Antiplatelet agents included, but were not limited to: acetylsalicylic acid (aspirin), adenosine reuptake inhibitors (dipyridamole), adenosine diphosphate receptor inhibitors (ticlopidine and clopidogrel), phosphodiesterase 3 inhibitors (cilostazol), P2Y₁₂ antagonists (prasugrel, ticagrelor, cangrelor, elinogrel), glycoprotein IIb/IIIa inhibitors (abciximab, eptifibatide, tirofiban, defibrotide), and sulfinpyrazone.
Types of outcome measures
- Myocardial infarction (nonfatal or fatal)
- Stroke (nonfatal or fatal)
- All-cause mortality
- Cardiovascular mortality
- Bleeding-related death
- Major bleeding
- Minor bleeding
- Haemorrhagic stroke
- End-stage kidney disease (ESKD)
- Kidney transplant graft loss
- Transplant rejection
- Dialysis vascular outcomes (failure, early thrombosis, loss of unassisted patency, failure to attain suitability for dialysis, and need for access intervention)
- Treatment withdrawal.
- End of treatment SCr
- End of treatment proteinuria.
Search methods for identification of studies
A systematic and comprehensive literature search was carried out to identify eligible RCTs. There was no language restriction.
We searched the Cochrane Renal Group's Specialised Register (to 24 January 2011) through contact with the Trials Search Co-ordinator using search terms relevant to this review.
The Cochrane Renal Group’s Specialised Register contains studies identified from:
- Quarterly searches of the Cochrane Central Register of Controlled Trials CENTRAL
- Weekly searches of MEDLINE OVID SP
- Handsearching of renal-related journals and the proceedings of major renal conferences
- Searching of the current year of EMBASE OVID SP
- Weekly current awareness alerts for selected renal journals
- Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.
Studies contained in the Specialised Register are identified through search strategies for CENTRAL, MEDLINE, and EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies as well as a list of handsearched journals, conference proceedings and current awareness alerts are available in the Specialised Register section of information about the Cochrane Renal Group.
See Appendix 1 for search terms used in strategies for this review.
Data collection and analysis
Selection of studies
All RCTs enrolling participants with CKD were considered as well as studies in broader populations in which outcome data for subgroups with CKD could be disaggregated. Based on the search strategy described, we identified titles and abstracts that were potentially relevant to this systematic review. Three independent authors (MR, LDM, SP) screened the titles and abstracts and selected those that met the inclusion criteria. Discrepancies in selection were resolved by discussion or by the review of an experienced arbitrator (GFMS). Studies reported in non-English language journals were translated before assessment.
Data extraction and management
Three authors independently read the full text of extracted articles and included studies that met the inclusion criteria. Where more than one publication of one study existed, reports were grouped together and the publication with the most complete data was used in the analyses.
The same independent authors used standardised data forms to extract data on:
- Study design
- Participants: baseline characteristics including age, gender, race, diabetic status (proportion with diabetes), hypertension status (proportion with hypertension), smoking status (proportion of smokers), visceral obesity (proportion with visceral obesity as defined by authors), previous cardiovascular events (proportion with existing cardiovascular disease), and stage of CKD (dialysis, predialysis, transplant)
- Interventions and comparisons: antiplatelet agent, dose and route of administration, duration of treatment
- Outcomes: as listed in Types of outcome measures.
Assessment of risk of bias in included studies
The risk of bias in included studies was formally assessed by looking at standard quality domains using the risk of bias assessment tool (Higgins 2011). The items assessed were sequence generation, allocation concealment, blinding (participants, investigators, and outcome assessors), completeness of outcome data, selective reporting and other potential sources of bias. We made explicit judgements (Appendix 2) regarding whether studies were at high risk of bias, according to criteria in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We explored the impact of bias by undertaking sensitivity analyses.
Measures of treatment effect
For dichotomous outcomes (e.g. such as death, cardiovascular events), results were expressed as risk ratio (RR) with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the effects of treatment (e.g. creatinine clearance (CrCl), GFR, SCr, proteinuria), the mean difference (MD) and its 95% CI was used. The final results are presented in International System (SI) units. When crude event data were not reported by investigators, available reported risk estimates and their 95% CIs were included in meta-analyses (CURE Study 2007; PLATO Study 2010).
Dealing with missing data
Where possible, data for each outcome of interest were evaluated, regardless of whether the analysis was based on intention-to-treat. In particular, dropout rates were investigated and reported in detail, including dropout due to discontinuation of study drug, treatment failure, death, withdrawal of consent, or loss to follow-up. Corresponding authors of all large studies with broader inclusion were contacted to obtain data for the subgroup of CKD.
Assessment of heterogeneity
We tested for heterogeneity using the Cochran Q test, which follows a Chi² distribution with n-1 degrees of freedom, with an alpha of < 0.10 used for statistical significance. The extent of heterogeneity was assessed with I², which ranges between 0% and 100% and expresses the proportion of between group variability that is attributable to heterogeneity rather than chance (Higgins 2003). I² values above 75% are typically held to signify extreme heterogeneity, whereas 25% and 50% correspond to low and medium levels of heterogeneity, respectively.
Assessment of reporting biases
We evaluated asymmetries in the inverted funnel plots (i.e. for systematic differences in the effect sizes between more precise and less precise studies). There are many potential explanations for why an inverted funnel plot may be asymmetric, including chance, heterogeneity, publication and reporting bias (Sterne 2011). Insufficient data were available to evaluate the robustness of the results according to publication, namely, publication as full manuscript in a peer reviewed journal versus studies published as abstracts/text/letters/editorials and publication.
Data were pooled using the random-effects model. The GRADE approach developed by Grades of Recommendation, Assessment, Development and Evaluation Working Group (GRADE Working Group) was used for evaluating the quality of evidence for outcomes to be reported. Based on the GRADE approach, the quality of a body of evidence, in terms of the extent to which one can be confident that an estimate of effect or association is close to the quantity of specific interest, was defined. Quality of a body of evidence involves consideration of within-study risk of bias (methodological quality), directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias ( Table 1). Factors that might decrease the quality level of a body of evidence were considered ( Table 2).
Subgroup analysis and investigation of heterogeneity
Heterogeneity was explored using subgroup analyses according to the following parameters (where sufficient numbers of studies were available):
- Population characteristics
- Stage of CKD (pre-dialysis, dialysis, transplant)
- Presence or absence of comorbidities (diabetes, hypertension, dyslipidaemia, smoking, obesity, family history of cardiovascular disease, baseline cardiovascular disease); percentage of patients with these comorbidities in each study
- Mean systolic blood pressure (BP) (below 140 mm Hg versus 140 mm Hg or above)
- Ethnicity (proportion white)
- Presence or absence of previous cardiovascular events (e.g. primary versus secondary prevention)
- Time on dialysis (fewer than three years versus three years or more) and modalities of dialysis (haemodialysis versus peritoneal dialysis)
- Time with a functioning transplant (fewer than three years versus three years or more).
- Intervention characteristics
- Types, doses and route of administration of the antiplatelet agents
- Duration of intervention (less than six months, six to 12 months, greater than 12 months).
We performed univariate meta-regression according to previously described methods when sufficient studies were identified (Palmer 2007). Univariate meta-regression was conducted using SAS (SAS Institute Inc, Cary, NC; Release 9·1, 2002-2003).
Sensitivity analyses were undertaken to explore the robustness of findings to key decisions in the review process. We assessed the risks of mortality, nonfatal and fatal myocardial infarction, and major bleeding only including studies with adequate allocation concealment, or at low risk of bias due to completeness of follow-up. Insufficient data were available to perform indirect comparisons of antiplatelet agent versus antiplatelet agent (Song 2003).
Description of studies
Results of the search
Electronic searching of MEDLINE, EMBASE, CENTRAL and the Cochrane Renal Group's Specialised Register (24 January 2011) identified 1075 citations (Figure 1). Searching of reference lists (8), previous systematic reviews (4), trials' registries (2), and data received from investigators (4) provided data for people with CKD from 18 other studies. We removed 64 duplicate citations to screen 1029 citations by title and abstract. Of these, 847 citations were excluded because they were not original investigations (reviews, commentaries, editorials), not RCTs, not appropriate interventions, had irrelevant outcomes, were animal studies or were conducted in children, or were not in people with CKD. We then assessed 182 full text articles for eligibility. Of these, 102 were excluded: 3 were commentaries, 16 were not RCTs, 28 were not appropriate interventions, 40 did not provide outcome data for participants with CKD, and 15 did not provide outcome data relevant to this review. Two ongoing studies (3 citations) were identified and will be assessed for inclusion in a future update of this review (FAVOURED Trial; NCT01198379).
|Figure 1. Study flow diagram; study identification and selection process|
We included 50 studies (27,139 participants) published in 77 reports; 44 studies (21,460 participants) compared an antiplatelet agent with placebo or no treatment (Abdul-Rahman 2007; Anderson 1974; Andrassy 1974; CHARISMA Study 2009; Cheng 1998; CREDO Study 2008; Creek 1990; CURE Study 2007; Dember 2008; Dixon 2009; Dodd 1980; Donadio 1984; Ell 1982; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; ETDRS 1992; Fiskerstrand 1985; Frascà 1997; Ghorbani 2009; Giustina 1998; Gonzalez 1995; Gröntoft 1985; Gröntoft 1998; Harter 1979; HOT Study 2010; IMPACT II Study 1997; Kaegi 1974; Kaufman 2003; Khajehdehi 2002; Kobayashi 1980; Kooistra 1994; Michie 1977; Middleton 1992; PRISM-PLUS Study 2002; PURSUIT Study 1998; Quarto Di Palo 1991; RAPPORT Study 1998; Schulze 1990; Sreedhara 1994; STOP Study 1995; Taber 1992; UK-HARP-I Study 2005; Zäuner 1994) and six studies (5679 participants) compared an antiplatelet agent with another antiplatelet agent (Frascà 1986; Kauffmann 1980; Ogawa 2008; PLATO Study 2010; TARGET Study 2001; TRITON-TIMI 38 2007).The overall characteristics of the included studies are provided in Characteristics of included studies. Information for three studies (1238 participants: Creek 1990; Ell 1982; Middleton 1992) including two internal study reports (Creek 1990; Middleton 1992) were only available in a previously published meta-analysis of antiplatelet agents (ATT 2002). For three studies (94 participants), the most complete data were provided in published conference proceedings (Dodd 1980; Gonzalez 1995; Taber 1992). One study was reported in German (Schulze 1990).
Antiplatelet versus placebo or no treatment studies
We included 44 studies comparing an antiplatelet to placebo or no treatment that were published between 1974 and 2010 (21,460 participants). The cohort size ranged from 16 to 4087 participants (median 100 participants). Data were available for the subgroup with CKD within a broader population in 12 studies (15,782 participants: CHARISMA Study 2009; CREDO Study 2008; CURE Study 2007; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; ETDRS 1992; HOT Study 2010; IMPACT II Study 1997; PRISM-PLUS Study 2002; PURSUIT Study 1998; RAPPORT Study 1998). The duration of follow-up ranged from one to 61 months (median nine months).
We identified 19 studies that were conducted in people with CKD not yet requiring dialysis (16,065 participants: CHARISMA Study 2009; Cheng 1998; CREDO Study 2008; CURE Study 2007; Donadio 1984; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; ETDRS 1992; Frascà 1997; Giustina 1998; Gonzalez 1995; HOT Study 2010; IMPACT II Study 1997; Khajehdehi 2002; PRISM-PLUS Study 2002; PURSUIT Study 1998; RAPPORT Study 1998; Zäuner 1994). A further 21 studies were in people on dialysis (4820 participants: Abdul-Rahman 2007; Andrassy 1974; Creek 1990; Dember 2008; Dixon 2009; Dodd 1980; Ell 1982; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Gröntoft 1998; Harter 1979; Kaegi 1974; Kaufman 2003; Kobayashi 1980; Kooistra 1994; Michie 1977; Middleton 1992; Sreedhara 1994; STOP Study 1995; Taber 1992). Three studies enrolled kidney transplant recipients (137 participants: Anderson 1974; Quarto Di Palo 1991; Schulze 1990) and one study (UK-HARP-I Study 2005; 448 participants) included people with CKD, transplant recipients and participants on dialysis.
In the 19 studies in people with CKD, two were in those with acute coronary syndromes (4698 participants: CURE Study 2007; PRISM-PLUS Study 2002); five enrolled participants with acute or stable coronary artery disease undergoing percutaneous intervention (2243 participants: EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; IMPACT II Study 1997; RAPPORT Study 1998); and one study enrolled people undergoing elective percutaneous coronary intervention (CREDO Study 2008).
Five studies enrolled people with diabetic kidney disease (2358 participants: CHARISMA Study 2009; ETDRS 1992; Giustina 1998; Gonzalez 1995; Khajehdehi 2002); four were in individuals with glomerulonephritis (119 participants: Cheng 1998; Donadio 1984; Frascà 1997; Zäuner 1994) and one study evaluated 3619 people with hypertension and CKD in a larger primary prevention study (HOT Study 2010).
The mean age of participants ranged from 30 to 75 years, with participants in studies of acute and stable coronary artery disease tending to be older (59 to 75 years) than those on dialysis (43 to 62 years) or with glomerulonephritis (30 to 45 years). The proportion of people with diabetes in the included studies was between 0% and 20% in two studies (HOT Study 2010; UK-HARP-I Study 2005); 20% to 70% in 13 studies (Abdul-Rahman 2007; CREDO Study 2008; CURE Study 2007; Dember 2008; Dixon 2009; EPIC Study 1994; EPISTENT Study 1998; Ghorbani 2009; IMPACT II Study 1997; Kaufman 2003; PRISM-PLUS Study 2002; PURSUIT Study 1998; RAPPORT Study 1998); and 100% in five studies (CHARISMA Study 2009; ETDRS 1992; Giustina 1998; Gröntoft 1998; Khajehdehi 2002).
In the 44 studies that compared an antiplatelet agent with placebo or no treatment, the interventions included:
- Aspirin or dipyridamole
- Clopidogrel and aspirin (1 study, 200 participants: Kaufman 2003)
- Glycoprotein IIb/IIIa inhibitors
- Tirofiban (1 study, 611 participants: PRISM-PLUS Study 2002)
- Defibrotide (1 study, 20 participants: Frascà 1997)
Non-randomised co-interventions were reported in 21 studies and included: angiotensin-converting enzyme inhibitors (2 studies, 89 participants: Cheng 1998; Gonzalez 1995); anticoagulation (2 studies, 873 participants: Kaegi 1974; STOP Study 1995); aspirin (4 studies, 6569 participants: CHARISMA Study 2009; CREDO Study 2008; CURE Study 2007; RAPPORT Study 1998); aspirin and heparin (5 studies, 5041 participants: EPIC Study 1994; EPILOG Study 1997; IMPACT II Study 1997; PRISM-PLUS Study 2002; PURSUIT Study 1998); aspirin, heparin and ticlopidine (1 study, 368 participants: EPISTENT Study 1998); immunosuppression (2 studies, 47 participants: Anderson 1974; Frascà 1997); avoidance of hypertension (1 study, 92 participants: Andrassy 1974); hypoglycaemic treatment and diet (1 study, 30 participants: Giustina 1998); and symptomatic therapy (1 study, 18 participants: Zäuner 1994). Two studies administered aspirin with either a BP target (3619 participants: HOT Study 2010) or simvastatin (448 participants: UK-HARP-I Study 2005) in a two-by-two factorial design.
Vascular access studies
We identified 21 studies that reported outcomes for dialysis vascular access in 4826 participants. Generally, these studies were small; only four studies included more than 500 participants (Dember 2008; Dixon 2009; Middleton 1992; STOP Study 1995), and nine studies enrolled fewer than 100 participants (Abdul-Rahman 2007; Andrassy 1974; Ell 1982; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Harter 1979; Kaegi 1974; Michie 1977). Ticlopidine was most commonly administered (6 studies, 760 participants: Creek 1990; Ell 1982; Fiskerstrand 1985; Gröntoft 1985; Gröntoft 1998; Kobayashi 1980) followed by aspirin (5 studies, 331 participants: Abdul-Rahman 2007; Andrassy 1974; Harter 1979; Kooistra 1994; Sreedhara 1994). The combination of dipyridamole and aspirin was prescribed to 1575 participants in three studies (Dixon 2009; Middleton 1992; Sreedhara 1994); two studies each evaluated clopidogrel (970 participants: Dember 2008; Ghorbani 2009) and sulphinpyrazone (78 participants: Kaegi 1974; Michie 1977); and single studies assessed dipyridamole (23 participants: Sreedhara 1994), picotamide (811 participants: STOP Study 1995) and the combination of clopidogrel and aspirin (200 participants: Kaufman 2003).
Studies evaluated treatment to maintain patency of an arteriovenous fistula (6 studies, 1259 participants: Andrassy 1974; Dember 2008; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Kooistra 1994); graft (4 studies, 975 participants: Dixon 2009; Kaufman 2003; Sreedhara 1994; Taber 1992); fistula or graft (2 studies, 1069 participants: Gröntoft 1998; STOP Study 1995); fistula or shunt (2 studies, 68 participants: Dodd 1980; Ell 1982); graft or shunt (1 study, 107 participants: Kobayashi 1980); shunt (3 studies, 1009 participants: Harter 1979; Kaegi 1974; Middleton 1992); fistula, graft or shunt (1 study, 16 participants: Michie 1977) or central venous catheter (1 study, 38 participants: Abdul-Rahman 2007). Antiplatelet agents were administered at the time of access surgery in 11 studies (2215 participants: Andrassy 1974; Dember 2008; Dixon 2009; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Gröntoft 1998; Harter 1979; Michie 1977; Sreedhara 1994; Taber 1992); of these, six studies started treatment one to two days before surgery (Andrassy 1974; Fiskerstrand 1985; Gröntoft 1985; Michie 1977; Sreedhara 1994; Taber 1992); two studies began antiplatelet treatment from seven to 10 days before access creation (Ghorbani 2009; Gröntoft 1998); two began treatment one to two days after surgery (Dember 2008; Dixon 2009); and one study commenced therapy one month after access surgery (Harter 1979). The duration of intervention varied from one month to five years, with a median of five months (interquartile range two to six months). Two recent large studies stratified randomisation according to the position of the dialysis access: forearm versus upper arm (Dember 2008); or other (Dixon 2009).
Antiplatelet versus antiplatelet studies
Six studies (5679 participants) compared an antiplatelet drug with a second antiplatelet drug in people with CKD and were published between 1980 and 2010. Since 2005, data for the subgroup of participants with CKD in larger studies of at-risk patients have become available for three studies (5517 participants: PLATO Study 2010; TARGET Study 2001; TRITON-TIMI 38 2007). We received unpublished data for the subgroup of participants with CKD in two larger studies of people scheduled to undergo percutaneous coronary intervention (TARGET Study 2001; TRITON-TIMI 38 2007). Four studies (5557 participants) enrolled people with CKD; three studies (5517 participants) enrolled people with acute or stable coronary artery disease undergoing percutaneous intervention; one the study (40 participants) enrolled people with diabetic kidney disease (Ogawa 2008). Two studies evaluated antiplatelet agents in 112 kidney transplant recipients (Frascà 1986; Kauffmann 1980). Study duration ranged from four to 48 months (median 12 months).
The mean age was available in four studies and ranged from 34 years in a study of kidney transplant recipients (Frascà 1986) to 74 years in people with acute coronary syndromes undergoing percutaneous coronary intervention (PLATO Study 2010; TRITON-TIMI 38 2007). People with diabetes were excluded from Frascà 1986, but made up approximately one-third of participants in PLATO Study 2010 and TRITON-TIMI 38 2007; all participants in the Ogawa 2008 study had been diagnosed with diabetes.
In the studies that compared an antiplatelet with another antiplatelet, interventions included:
- Aspirin versus sarpogrelate (40 participants; Ogawa 2008)
- Dipyridamole versus defibrotide (80 participants; Frascà 1986)
- Aspirin versus dipyridamole (42 participants; Kauffmann 1980)
- Thienopyridine versus thienopyridine
- Glycoprotein IIb/IIIa inhibitor
- Abciximab versus tirofiban (790 participants; TARGET Study 2001).
Risk of bias in included studies
The risk of bias in the included studies is summarised in Figure 2 and results for individual studies are reported in the Characteristics of included studies table. Overall, random sequence generation was unclear or not adequate in 98% of studies. Data follow-up were generally complete in half of the studies, whereas selective reporting was likely in 60% of studies.
|Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies|
Allocation concealment was adequate in 19 studies (38%), not adequate in one (allocation to treatment according to medical record number; 2%), and unclear in the remaining 30 studies (60%).
Participants, investigators, and outcome assessors were blinded in 20 studies (40%), were not all blinded in four (8%), and blinding was unclear in the remaining 26 studies (52%).
Incomplete outcome data
Data follow-up was complete in 50% of the studies, incomplete in 12 (24%) and unclear in the remaining 13 studies (26%).
Reporting of outcomes was complete in 20 studies (40%), clearly selective in three (6%), and unclear in 27 studies (54%).
Other potential sources of bias
Other potential sources of bias were identified in 15 studies including a change in the analysis plan after recruitment (Kaegi 1974) full study reports were not available for six studies (Creek 1990; Dodd 1980; Ell 1982; Gonzalez 1995; Middleton 1992; Taber 1992); and/or data not available for meta-analysis (Frascà 1997); baseline differences in treatment groups (Harter 1979); and early termination of one or more arms of the study (6 studies: Dember 2008; EPILOG Study 1997; Kaufman 2003; PRISM-PLUS Study 2002; PURSUIT Study 1998; UK-HARP-I Study 2005).
Effects of interventions
Antiplatelet versus placebo or no treatment
Summary effects of antiplatelet agents versus placebo or no treatment for myocardial infarction, all-cause mortality, and major bleeding are summarised in Table 2.
Fatal or nonfatal myocardial infarction
Seventeen studies (14,451 participants) reported effects of antiplatelet treatment on fatal or nonfatal myocardial infarction. Antiplatelet therapy reduced the risk of myocardial infarction by 13% ( Analysis 1.1: RR 0.87, 95% CI 0.76 to 0.99). There was no evidence of significant heterogeneity in this meta-analysis (Tau² = 0.00; Chi² = 14.47, df = 14 (P = 0.42); I² = 3%).
Subgroup analysis for myocardial infarction - stratified by stage of CKD
There was no difference in fatal or nonfatal myocardial infarction based on stage of CKD. The RR in people with earlier stages of CKD was 0.84 (95% CI 0.70 to 0.99; 10 studies, 11,074 participants) (CHARISMA Study 2009; CREDO Study 2008; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; ETDRS 1992; HOT Study 2010; IMPACT II Study 1997; PURSUIT Study 1998; RAPPORT Study 1998) whereas the risk in people requiring dialysis was 0.82 (95% CI 0.47 to 1.42; 6 studies, 2929 participants) (Creek 1990; Dember 2008; Dixon 2009; Ell 1982; Kaufman 2003; STOP Study 1995; P = 0.94).
Subgroup analysis for myocardial infarction - stratified by antiplatelet type
There was no difference between antiplatelet types, suggesting the overall risk for myocardial infarction was the most robust (test for subgroup differences P = 0.20). The RR of myocardial infarction was 0.68 (95% CI 0.46 to 0.99) for aspirin (3 studies, 4252 participants: ETDRS 1992; HOT Study 2010; UK-HARP-I Study 2005), 0.83 (95% CI 0.57 to 1.21) for thienopyridines (5 studies, 3689 participants: CHARISMA Study 2009; CREDO Study 2008; Creek 1990; Dember 2008; Ell 1982) and 0.93 (95% CI 0.81 to 1.07) for glycoprotein IIb/IIIa inhibitors (6 studies, 4850 participants: EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; IMPACT II Study 1997; PURSUIT Study 1998; RAPPORT Study 1998). There were insufficient numbers of studies that were clearly primary prevention studies to enable analysis of whether the type of intervention (primary versus secondary prevention) was an effect modifier on the risk of myocardial infarction.
Among studies comparing antiplatelet agents with placebo/no treatment, less precise (smaller studies) with smaller or no treatment effect were not available on the funnel plot (Figure 3).
|Figure 3. Funnel plot of comparison of antiplatelet agents versus control for the outcome of fatal or nonfatal myocardial infarction|
Fatal or nonfatal stroke
Eleven studies (9544 participants) reported the risk of fatal or nonfatal stroke. Antiplatelet therapy did not alter the risk of stroke ( Analysis 1.2: RR 1.00, 95% CI 0.58 to 1.72). There was no significant heterogeneity in this meta-analysis (Tau² = 0.23; Chi² = 14.15, df = 8 (P = 0.08); I² = 43%).
Six studies (6044 participants) reported the risk of haemorrhagic stroke ( Analysis 1.3: RR 1.22, 95% CI 0.69 to 2.17). There was no heterogeneity in this meta-analysis (Tau² = 0.00; Chi² = 1.38, df = 5 (P = 0.93); I² = 0%).
All-cause mortality was reported in 29 studies (16,152 participants). Antiplatelet treatment did not lower all-cause mortality ( Analysis 1.4: RR 0.95, 95% CI 0.83 to 1.07). There was no heterogeneity in this meta-analysis (Tau² = 0.01; Chi² = 22.99, df = 20 (P = 0.29); I² = 13%). When data from CURE Study 2007 (4087 participants), which were available only as a risk estimate and its variance were included in this meta-analysis, the risk of all-cause mortality remained similar between antiplatelet and control (30 studies, 20,239 participants: RR 0.93, 95% CI 0.81 to 1.06), without significant heterogeneity.
Subgroup analysis for mortality - type of antiplatelet treatment
In subgroup analysis for the effect of different antiplatelet types on mortality, there was no evidence of interaction between subgroups (P = 0.46). In five aspirin studies (4340 participants), the RR of all-cause mortality was 0.88 (95% CI 0.61 to 1.27) (ETDRS 1992; HOT Study 2010; Khajehdehi 2002; Sreedhara 1994; UK-HARP-I Study 2005); in 10 studies of thienopyridines (8210 participants), including CURE Study 2007, the risk of all-cause mortality was 1.10 (95% CI 0.80 to 1.51) (CHARISMA Study 2009; Cheng 1998; CREDO Study 2008; Creek 1990; CURE Study 2007; Dember 2008; Ell 1982; Ghorbani 2009; Gröntoft 1998; Kobayashi 1980), and in six studies of glycoprotein IIb/IIIa inhibitors (4849 participants) the risk of all-cause mortality was 0.83 (95% CI 0.60 to 1.16) (EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; IMPACT II Study 1997; PURSUIT Study 1998; RAPPORT Study 1998).
Subgroup analysis for mortality - stage of CKD
No differences were seen in all-cause mortality among 15,297 people with earlier stages of CKD (15 studies: RR 0.96, 95% CI 0.82 to 1.14) (CHARISMA Study 2009; Cheng 1998; CREDO Study 2008; CURE Study 2007; Donadio 1984; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; ETDRS 1992; HOT Study 2010; IMPACT II Study 1997; Khajehdehi 2002; PURSUIT Study 1998; RAPPORT Study 1998; Zäuner 1994) and 4363 people requiring dialysis (13 studies: RR 0.82, 95% CI 0.63 to 1.06; P = 0.30) (Creek 1990; Dember 2008; Dixon 2009; Ell 1982; Ghorbani 2009; Gröntoft 1998; Kaegi 1974; Kaufman 2003; Kobayashi 1980; Michie 1977; Middleton 1992; Sreedhara 1994; STOP Study 1995).
Cardiovascular mortality data were available in 18 studies (9337 participants) (CHARISMA Study 2009; Cheng 1998; CREDO Study 2008; Creek 1990; Donadio 1984; Ell 1982; ETDRS 1992; Gröntoft 1998; HOT Study 2010; Kaegi 1974; Khajehdehi 2002; Michie 1977; Quarto Di Palo 1991; STOP Study 1995; UK-HARP-I Study 2005; Zäuner 1994). Overall, antiplatelet agents did not reduce cardiovascular mortality ( Analysis 1.5: RR 0.87, 95% CI 0.65 to 1.15) without evidence of significant heterogeneity (Tau² = 0.06; Chi² = 13.26, df = 9 (P = 0.15); I² = 32%). When we included data from CURE Study 2007 that provided a risk estimate and its variance for cardiovascular mortality, antiplatelet agents were not better than control for reducing this outcome (19 studies, 13,424 participants: RR 0.89, 95% CI 0.70 to 1.12), without significant heterogeneity in the analysis (Chi² = 14.1, df = 11 (P = 0.72); I² = 0%). Studies of glycoprotein IIb/IIIa inhibitors did not provide data for cardiovascular mortality.
Major bleeding events were: retroperitoneal, intra-articular, intra-ocular, intracranial or intracerebral haemorrhage, gastrointestinal bleeding, bleeding that was fatal, life-threatening, disabling or required transfusion, corrective surgery or hospitalisation, with or without a fall in haemoglobin level of at least 2 g/dL (Abdul-Rahman 2007; CHARISMA Study 2009; Dixon 2009; Ghorbani 2009; HOT Study 2010; Kaegi 1974; Kobayashi 1980; PRISM-PLUS Study 2002; UK-HARP-I Study 2005) or melaena (Andrassy 1974). Minor bleeding events were usually described as not serious or significant (CHARISMA Study 2009; CREDO Study 2008; CURE Study 2007; Dixon 2009; HOT Study 2010); epistaxis, ecchymoses, or bruising (Andrassy 1974, UK-HARP-I Study 2005); not requiring transfusion, hospitalisation, or an event-related study visit (CHARISMA Study 2009; Kaufman 2003); or bleeding from cannulation sites or haematuria (Gröntoft 1985).
Twenty six studies (15,992 participants) that provided event data for major bleeding (Abdul-Rahman 2007; Andrassy 1974; CHARISMA Study 2009; CREDO Study 2008; Creek 1990; Dember 2008; Dixon 2009; Ell 1982; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; Ghorbani 2009; Harter 1979; HOT Study 2010; IMPACT II Study 1997; Kaegi 1974; Khajehdehi 2002; Kobayashi 1980; Michie 1977; Middleton 1992; PRISM-PLUS Study 2002; PURSUIT Study 1998; Quarto Di Palo 1991; RAPPORT Study 1998; STOP Study 1995; UK-HARP-I Study 2005). Antiplatelet agents significantly increased major bleeding ( Analysis 1.7: RR 1.35, 95% CI 1.10 to 1.65). There was no significant heterogeneity in this analysis (Tau² = 0.02; Chi² = 20.38, df = 18 (P = 0.31); I² = 12%). Addition of the risk estimate for major bleeding reported in CURE Study 2007 (4087 participants) to the meta-analysis resulted in a similar risk of bleeding (RR 1.33, 95% CI 1.09 to 1.64) without evidence of heterogeneity.
Subgroup analysis for major bleeding - stratified by stage of CKD
There was no difference in major bleeding based on stage of CKD. The RR of major bleeding was 1.45 (95% CI 1.18 to 1.8) in people with CKD (12 studies) (CHARISMA Study 2009; CREDO Study 2008; CURE Study 2007; EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; HOT Study 2010; IMPACT II Study 1997; Khajehdehi 2002; PRISM-PLUS Study 2002; PURSUIT Study 1998; RAPPORT Study 1998) and 0.93 (95% CI 0.55 to 1.57; P = 0.12) in people on dialysis (13 studies) (Abdul-Rahman 2007; Andrassy 1974; Creek 1990; Dember 2008; Dixon 2009; Ell 1982; Ghorbani 2009; Harter 1979; Kaegi 1974; Kobayashi 1980; Michie 1977; Middleton 1992; STOP Study 1995). The Quarto Di Palo 1991 study in kidney transplant recipient was not included in the subgroup analysis.
Subgroup analysis for major bleeding - stratified by antiplatelet type
There was no difference in major bleeding based on antiplatelet type. The RR of major bleeding were 1.34 (95% CI 0.71 to 2.55) in seven aspirin studies (Abdul-Rahman 2007; Andrassy 1974; Ell 1982; Harter 1979; HOT Study 2010; Khajehdehi 2002; UK-HARP-I Study 2005); 1.27 (95% CI 0.85 to 1.91) in seven thienopyridine studies (CHARISMA Study 2009; CREDO Study 2008; Creek 1990; CURE Study 2007; Dember 2008; Ghorbani 2009; Kobayashi 1980); and 1.45 (95% CI 1.04 to 2.04) in seven glycoprotein IIb/IIIa studies (EPIC Study 1994; EPILOG Study 1997; EPISTENT Study 1998; IMPACT II Study 1997; PRISM-PLUS Study 2002; PURSUIT Study 1998; RAPPORT Study 1998; P = 0.88).
Numbers of major bleeding events in studies of dual antiplatelet therapy (Dixon 2009; Khajehdehi 2002; Middleton 1992) were insufficient to determine whether dual antiplatelet therapy was an effect modifier on major bleeding in the available studies.
Antiplatelet agents increased the risk of minor bleeding ( Analysis 1.8 (18 studies, 13,106 participants): RR 1.54, 95% CI 1.26 to 1.90). There was significant heterogeneity in this analysis (Tau² = 0.06; Chi² = 35.12, df = 13 (P = 0.0008); I² = 63%). When the risk estimate for major bleeding from CURE Study 2007 was included in the analysis, a similar risk of minor bleeding with antiplatelet agents was identified (RR 1.48, 95% CI 1.12 to 1.97) with persistent heterogeneity (Chi² = 33.1 (P = 0.003); I² = 58%).
Subgroup analysis for minor bleeding to explore heterogeneity
To explore heterogeneity, we conducted subgroup analyses for minor bleeding according to study and population characteristics as defined a priori in the study protocol. There was no difference in risk of minor bleeding between subgroups based on stage of CKD, presence of cardiovascular disease at baseline, duration of intervention, age or gender of participants, or presence of hypertension or diabetes (P > 0.15 for all). Subgroup analyses based on duration of dialysis or transplantation were not possible due to insufficient numbers of studies.
The risk of ESKD was not altered with antiplatelet therapy ( Analysis 1.9 (8 studies, 825 participants): RR 0.96, 95% CI 0.67 to 1.37) without evidence of significant heterogeneity (Tau² = 0.02; Chi² = 4.56, df = 4 (P = 0.34); I² = 12%).
There was no difference between antiplatelet agents and control for the risk doubling of SCr ( Analysis 1.10 (2 studies, 126 participants): RR 0.43, 95% CI 0.12 to 1.47).
The risk of kidney transplant graft loss was not altered by antiplatelet treatment ( Analysis 1.11 (2 studies, 91 participants): RR 1.08, 95% CI 0.58 to 2.01) without evidence of heterogeneity (Tau² = 0.00; Chi² = 0.16, df = 1 (P = 0.69); I² = 0%). Transplant rejection was not altered by antiplatelet therapy ( Analysis 1.12 (2 studies, 97 participants): RR 0.95, 95% CI 0.77 to 1.19).
End of treatment CrCl was similar between antiplatelet and control treatment in two studies ( Analysis 1.13 (68 participants): MD -6.41 mL/min, 95% CI -19.94 to 7.12) with no significant heterogeneity evident in the analyses (Tau² = 61.98; Chi² = 2.01, df = 1 (P = 0.16); I² = 50%). There was no difference between treatment groups for the end of treatment proteinuria ( Analysis 1.14 (2 studies, 54 participants): MD -1.43 g/d, 95% CI -3.84 to 0.98) with high level heterogeneity in the analysis (Tau² = 2.91; Chi² = 26.53, df = 1 (P < 0.0001); I² = 96%) suggesting it was inappropriate to combine the results of these studies. No difference for this outcome was observed within each study individually.
Dialysis access outcomes
Dialysis access failure (thrombosis or loss of patency)
Overall, 14 studies (2608 participants) reported dialysis access failure (thrombosis or loss of patency) (Abdul-Rahman 2007; Andrassy 1974; Dember 2008; Dixon 2009; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Gröntoft 1998; Harter 1979; Kaegi 1974; Kaufman 2003; Kooistra 1994; Michie 1977; Sreedhara 1994). For all access types, antiplatelet therapy reduced access failure by 32% ( Analysis 1.15 (14 studies, 2608 participants): RR 0.68, 95% CI 0.54 to 0.84). There was significant heterogeneity in this analysis (Tau² = 0.05; Chi² = 22.82, df = 13 (P = 0.04); I² = 43%) which we explored using subgroup analysis by access type: fistula (7 studies, 1502 participants); shunt or graft (5 studies, 1102 participants); fistula or graft (1 study, 16 participants); or central venous catheter (1 study, 38 participants). In these analyses, antiplatelet agents (aspirin, ticlopidine, or clopidogrel) reduced fistula thrombosis or patency failure by 44% ( Analysis 1.15.1: RR 0.56, 95% CI 0.40 to 0.78), but not shunt or graft failure ( Analysis 1.15.2: RR 0.80, 95% CI 0.62 to 1.03). However, the effect estimate was statistically similar for this outcome in fistula studies compared with those evaluating shunts or grafts (P = 0.10 for subgroup interaction). Overall, there was no evidence of subgroup interaction based on access type across all types, suggesting the specific vascular access (fistula, graft, shunt, or central venous catheter) (test for subgroup differences Chi² = 3.58, P = 0.31, I² = 16.2%) was not an effect modifier for the treatment effects observed and indicating the overall effect estimate was the most appropriate. Exclusion of the single study of central venous catheter use did not meaningfully change the overall treatment effect in favour of antiplatelet therapy (RR 0.69, 95% CI 0.55 to 0.86).
Early access failure (within eight weeks of access creation)
Six studies (1365 participants) evaluated access outcomes at or before eight weeks after surgery (Andrassy 1974; Dember 2008; Fiskerstrand 1985; Ghorbani 2009; Gröntoft 1985; Gröntoft 1998). Antiplatelet treatment reduced the risk of early access thrombosis ( Analysis 1.16: RR 0.54, 95% CI 0.39 to 0.74). There was no significant heterogeneity in this analysis (Tau² = 0.02; Chi² = 5.64, df = 5 (P = 0.34); I² = 11%).
Loss of unassisted patency
Two studies (665 participants: Dixon 2009; Michie 1977) reported loss of unassisted patency in an analysis in which a single recent study provided 99% of the events (Dixon 2009). Antiplatelet treatment did not reduce loss of unassisted patency ( Analysis 1.17: RR 0.95, 95% 0.89 to 1.03) with no heterogeneity (Tau² = 0.00; Chi² = 0.22, df = 1 (P = 0.64); I² = 0%).
Failure to attain access suitability of dialysis (maturation)
Five studies (1503 participants) provided data for suitability of vascular access for dialysis as an outcome. The definitions of access suitability included: the ability to use the fistula for dialysis with two needles and maintain a blood flow rate ≥ 300 mL/min during eight of 12 dialysis sessions occurring during a 30 day suitability ascertainment period (Dember 2008); failure to use graft by week 12 in patients with catheter for access (Dixon 2009); fistula ceased to function (Gröntoft 1985); permanent shunt thrombosis (Harter 1979); and failure to develop adequate flow (Michie 1977). Antiplatelet therapy did not reduce the risk of failure to attain access suitability ( Analysis 1.18: RR 0.62, 95% CI 0.33 to 1.16). There was significant heterogeneity in this analysis (Tau² = 0.25; Chi² = 10.11, df = 4 (P = 0.04); I² = 60%) potentially due to the differences in definitions for this outcome in the available studies. An insufficient number of studies reporting this outcome precluded formal subgroup analysis to explore heterogeneity.
Need for intervention to attain patency or assist maturation
Four studies (3980 participants) reported the need for the intervention to attain patency or assist maturation described as surgical revision (Kaegi 1974), thrombectomy (Michie 1977), percutaneous intervention to restore patency or promote maturation (Dember 2008), or angioplasty (Dixon 2009). Antiplatelet therapy did not reduce the risk for the need for the intervention to attain patency or assist maturation ( Analysis 1.19: RR 0.79, 95% CI 0.54 to 1.15) with no significant heterogeneity (Tau² = 0.08; Chi² = 3.99, df = 3 (P = 0.26); I² = 25%).
There were no differences in all-cause hospitalisation between treatment groups in three studies ( Analysis 1.20 (3535 participants): RR 0.97, 95% CI 0.87 to 1.10) without evidence for heterogeneity (Tau² = 0.00; Chi² = 1.85, df = 2 (P = 0.40); I² = 0%). Cardiovascular hospitalisation was similar between antiplatelet treatment and control in the two available studies ( Analysis 1.21 (1526 participants): RR 0.93, 95% CI 0.76 to 1.14). There was significant heterogeneity in this analysis (Tau² = 0.42; Chi² = 6.99, df = 1 (P = 0.008); I² = 86%) potentially due to differences in adjudication of the outcome.
Withdrawal from treatment
Thirteen studies (2569 participants) reported premature withdrawal from treatment. There was no difference in the risk of withdrawn form treatment between antiplatelet agents and placebo or no treatment ( Analysis 1.22: RR 0.97, 95% CI 0.83 to 1.13) with no heterogeneity (Tau² = 0.00; Chi² = 9.71, df = 11 (P = 0.56); I² = 0%).
Subgroup analyses by methodological quality for all-cause mortality and major bleeding
Analyses, including only studies with adequate allocation concealment, blinding, or completeness of follow-up, specified a priori, showed similar results for treatment efficacy for mortality or harm from major bleeding.
Antiplatelet versus antiplatelet
Data for a thienopyridine (prasugrel or ticagrelor) compared with clopidogrel were available in two studies (4727 participants: TRITON-TIMI 38 2007; PLATO Study 2010) involving people with CKD. Where possible, data from PLATO Study 2010 were provided as a risk estimate (hazard ratio) together with its 95% CI and were combined with event data from TRITON-TIMI 38 2007 study. Unpublished data for people with CKD were also available for the comparison of a glycoprotein IIb/IIIa inhibitor (abciximab) versus a second glycoprotein IIb/IIIa inhibitor (tirofiban) in 790 people from TARGET Study 2001 at six months for myocardial infarction and 12 months for all-cause mortality.
Thienopyridine versus clopidogrel
Fatal or nonfatal myocardial infarction
TRITON-TIMI 38 2007 (1490 participants) provided data for nonfatal myocardial infarction comparing prasugrel with clopidogrel. There was no difference between treatment groups for this outcome ( Analysis 2.1: RR 0.78, 95% CI 0.58, 1.05).
Fatal or nonfatal stroke
PLATO Study 2010 (3237 participants) reported a risk of intracranial haemorrhage comparing ticagrelor with clopidogrel of 1.79 (95% CI 0.43 to 7.51).
Two studies (4727 participants) compared a newer generation thienopyridine with clopidogrel for effects on all-cause mortality. When results from these studies were combined, the analysis had a high level of heterogeneity (Chi² = 1747.7, df = 1, P < 0.0001; I² = 99%), suggesting a summary estimate was not appropriate. Individually, prasugrel versus clopidogrel (1490 participants: TRITON-TIMI 38 2007) resulted in a risk for mortality of 0.81 (95% CI 0.56 to 1.18) whereas ticagrelor versus clopidogrel (3237 participants: PLATO Study 2010) had a reported risk estimate of 0.72 (95% CI 0.58 to 0.89).
Cardiovascular mortality was available from TRITON-TIMI 38 2007 (1490 participants). The risk of cardiovascular mortality was similar between the prasugrel and clopidogrel groups ( Analysis 2.3: RR 1.35, 95% CI 0.87 to 2.10).
PLATO Study 2010 (3237 participants) reported the risk between ticagrelor and clopidogrel for fatal bleeding as 0.48 (95% CI 0.15 to 1.54).
Two studies comparing prasugrel or ticagrelor to clopidogrel in 4706 individuals with CKD provided data for major bleeding. Major bleeding was defined according to the Thrombolysis In Myocardial Infarction criteria for major bleeding (intracranial haemorrhage, clinically evident bleeding including imaging and a drop in the haemoglobin of ≥5 g/dL (TRITON-TIMI 38 2007) or according to the Platelet Inhibition and Patient Outcomes (PLATO) definition as bleeding that led to clinically significant disability (e.g. intra-ocular bleeding with permanent vision loss) or bleeding either associated with a drop in haemoglobin level of at least 3.0 g/dL but less than 5.0 g/dL or requiring transfusion of two to three units of red cells (PLATO Study 2010). Combining unpublished crude event data (TRITON-TIMI 38 2007) in one study ( Analysis 2.4) with a risk estimate provided in the other study (PLATO Study 2010) by meta-analysis resulted in a summary risk estimate for major bleeding (RR 1.12, 95% CI 0.90 to 1.39). There was no significant heterogeneity in this analysis (Chi² = 1.08, df = 1 (P < 0.30); I² = 8%).
TRITON-TIMI 38 2007 (1469 participants) compared prasugrel with clopidogrel and provided unpublished data for minor bleeding in people with CKD (RR 1.35, 95% CI 0.87 to 2.10; Analysis 2.5). Minor bleeding was defined as clinically evident bleeding including imaging and a fall in the haemoglobin of between 3 and 5 g/dL (TRITON-TIMI 38 2007).
Glycoprotein IIb/IIIa inhibitor versus glycoprotein IIb/IIIa inhibitor
Investigators of TARGET Study 2001, which compared a glycoprotein IIb/IIIa inhibitor (abciximab) with another glycoprotein IIb/IIIa inhibitor (tirofiban), provided unpublished data for 790 people with CKD for outcomes relevant to this review of myocardial infarction at six months and all-cause mortality at 12 months follow-up.
Fatal or nonfatal myocardial infarction
Sensitivity and subgroups analyses
Sensitivity and subgroup analyses were not possible when comparing one antiplatelet with another antiplatelet were not possible due to the insufficient number of available studies.
Summary of main results
This review indicated that antiplatelet therapy (aspirin, thienopyridines, or glycoprotein IIb/IIIa inhibitors) lowers the risk for myocardial infarction in people with CKD, but does not reduce total or cardiovascular mortality or stroke. Importantly, antiplatelet treatment increases major bleeding by 33% (including bleeding events that result in hospital admission, transfusion, or disability) and minor bleeding by 48%. There is currently insufficient available evidence to define clearly the role of antiplatelet treatment in primary prevention (preventing cardiovascular events in people without existing cardiovascular disease) in those with CKD.
Antiplatelet treatment started around the time of vascular access surgery was found to reduce vascular access thrombosis or patency failure by a third, but there was insufficient evidence to show that antiplatelet therapy improves dialysis access maturation or access suitability for dialysis. Overall, antiplatelet agents do not prevent ESKD or kidney transplant loss.
Direct comparisons of antiplatelet agents are limited to a few studies in which data for the subgroup of participants with CKD have been recently reported or provided. Currently, there are scant data to recommend that one antiplatelet agent is more efficacious than another in any clinical setting (primary prevention or secondary prevention), particularly for people with acute coronary syndromes or those undergoing percutaneous coronary interventions who frequently have coexistent CKD.
Overall completeness and applicability of evidence
While the analyses include data obtained from a comprehensive search and unpublished data from numerous investigators, particularly for cardiovascular events, the data were incomplete in several areas. First, data for transplant recipients were limited and provided by smaller and older studies, published between 1974 and 1991. A more recent study of aspirin included transplant recipients in addition to individuals with CKD and those requiring dialysis (UK-HARP-I Study 2005) but data for the transplant subgroup (133 participants) were not available and would have provided very few events for relevant clinical outcomes. Outcome data for kidney transplant recipients were restricted generally to transplant function or rejection in two studies, and information about major cardiovascular events were scarce. Further, no head-to-head comparisons of antiplatelet agents were available in kidney transplant recipients or for any antiplatelet agent in kidney transplant recipients with acute coronary syndromes. Further research is needed in these populations. Second, no data for cardiovascular mortality were available in studies of glycoprotein IIb/IIIa inhibitors administered in addition to standard therapy at the time of an acute coronary syndrome with or without percutaneous coronary intervention. The data for people with CKD in studies assessing glycoprotein IIb/IIIa inhibitors were disaggregated from within a larger, at-risk populations after publication of the primary studies. Accordingly, to date no studies have specifically evaluated these agents in people with CKD a priori. Third, data for the effects of antiplatelet agents on primary prevention of cardiovascular events in people with CKD were available only from a single study of aspirin in people with hypertension in which data for the subgroup with CKD have become available (HOT Study 2010). The results of this study are consistent with the overall findings of the review showing no clear benefit for mortality, a reduction in myocardial infarction, and a near doubling in the risk of major bleeding (RR 2.04, 95% CI 1.05 to 3.96).
Quality of the evidence
Although this review found consistent effect estimates for important clinical outcomes (myocardial infarction, mortality, and bleeding) in analyses that include approximately 15,000 people with CKD and between 500 to 1000 events, our conclusions must be considered more cautiously due to several potential limitations in the available data.
While the analyses include data obtained from a comprehensive search and unpublished data from numerous investigators particularly for cardiovascular events, selective reporting of outcomes may reduce the strength of our conclusions. Data for myocardial infarction in smaller studies with smaller treatment benefits were absent because these (less precise) studies did not systematically report cardiovascular events. Accordingly, selective outcome reporting reduced the reliability of this treatment effect (13% reduction) in both magnitude and direction, although the effect of bias could not be determined in the absence of all data for this outcome. The small proportion of studies reporting vascular access outcomes including only 2000 participants reduced the strength of evidence for antiplatelet agents on vascular access function and maturation. Only two thirds of such studies reported access failure or thrombosis, and only 10% to 20% reported on maturation and suitability for dialysis outcomes in these people. Overall, 60% of studies did not report adequate blinding, allocation concealment or random sequence generation, although sensitivity analyses did not find differences in treatment effects when analyses were restricted to studies of higher methodological quality, because lower quality studies tended to be smaller and contributed fewer events to analyses. In addition, the number of major bleeding events in studies of dual antiplatelet therapy were insufficient to determine in indirect evidence whether bleeding risk was increased with dual antiplatelet therapy compared with monotherapy. Data from studies that directly compared two antiplatelet agents against a single antiplatelet agent were absent.
Consistency of results
Our major findings, that antiplatelet agents reduce myocardial infarction but not cardiovascular death or overall mortality, and significantly increase major bleeding are strengthened by the consistency of findings across studies. Two thirds of studies reported all-cause mortality in over 15,000 participants and showed no treatment effect in all but one study (CHARISMA Study 2009). Notably, this study of clopidogrel and aspirin versus aspirin alone in people with diabetic kidney disease showed that participants allocated to clopidogrel experienced significantly higher mortality, although the reasons for this finding remain unclear and might be expected by chance across the 30 studies. Similarly, in analyses for cardiovascular mortality that included 19 studies and nearly 10,000 participants, only CHARISMA Study 2009 had a 95% CI that did not include '1' suggesting the null effect of antiplatelet agents on cause-specific mortality is robust. There was also very low heterogeneity in the summary estimate for myocardial infarction, although only 17/44 (˜40%) potentially eligible studies reported this outcome. Approximately two thirds of placebo/no treatment studies reported major bleeding events with a consistent risk across all contributing studies of over 15,000 participants and nearly 600 events. The highly variable definitions of major bleeding in the included studies, together with the relative lack of specific reporting on intracranial haemorrhage (in only eight studies), reduced the ability to weigh the relative benefits of treatment (reducing myocardial infarction) with the comprehensive potential risks of harm for people with CKD. The risks of minor bleeding varied among studies beyond chance alone and could not be explained by subgroup analyses that included analyses for age, gender, pre-existing comorbidities or time on dialysis, reducing the reliability of the effect estimate identified for this outcome.
Directness of evidence
Only six studies reported direct comparisons of two antiplatelet agents, and meta-analyses were generally not possible. The small number of studies that directly compared different agents (glycoprotein inhibitors in one study and thienopyridines in two studies) precluded indirect comparisons of the magnitude of effect of each drug class (although such evidence is of lower quality than head-to-head comparisons of antiplatelet treatments). Therefore, we sought to identify whether a specific antiplatelet agent was particularly beneficial (or harmful) and if treatment effects varied with stage of CKD (dialysis versus earlier stages of CKD) using prespecified subgroup analyses categorised by antiplatelet drug type. No differences in treatment effects or harms were found among subgroups (aspirin, thienopyridine, or glycoprotein IIb/IIIa inhibitor) suggesting the effects of antiplatelet treatment observed are applicable to all these antiplatelet types used in the studies.
Effect estimates for major treatment benefits and harms (mortality, myocardial infarction, and major bleeding) had narrow confidence intervals, increasing their certainty and strengthening the evidence within the review for these clinical events. Several outcomes, however, included both few participants and events, indicating the available evidence for benefits (and toxicities) of antiplatelet agents for these outcomes is of lower quality. These outcomes included bleeding-related death, fatal and nonfatal stroke, haemorrhagic stroke, ESKD, transplant function and rejection, dialysis vascular access maturation, and hospitalisation. Effect estimates for direct antiplatelet versus antiplatelet comparisons were also very imprecise.
Agreements and disagreements with other studies or reviews
The results of this review expand the available evidence for people with CKD including data for 21,000 participants. An earlier collaborative systematic overview of 287 RCTs of an antiplatelet drug versus control (130,000 participants) or of one antiplatelet treatment versus another (77,000 participants) in people at high risk of cardiovascular disease (acute or previous vascular disease or other predisposing condition) included 2632 people requiring haemodialysis (ATT 2002). This review found that antiplatelet therapy produced a 41% proportional reduction in serious vascular events in this population. However, only 99 vascular events and 46 major extracranial bleeds were available at the time of publication nearly a decade ago (2002), limiting the reliability of the conclusions drawn (ATT 2002). Data for people with earlier stages of CKD were not available in this earlier review and have only recently become more available. A recent systematic review of individual patient data for aspirin in the primary and secondary prevention of vascular disease did not provide specific analyses for individuals based on presence of CKD (ATT 2009).
Notably, our systematic review (that finds a 13% lowering of myocardial infarction, a one third increase in bleeding, and no benefit on mortality outcomes) differs from these two previous studies. We suggest that the benefits of antiplatelet agents on cardiovascular events may be smaller in people with CKD compared with other populations at risk of cardiovascular events, for whom reductions in cardiovascular mortality, myocardial infarction, and stroke are one sixth, one third, and one quarter respectively. The relatively reduced efficacy for antiplatelet agents on total mortality in CKD is potentially explained by the competing mechanisms for cardiovascular death in this population. Progressive kidney dysfunction is characterised by vascular stiffening and calcification, cardiomyopathy, hyperkalaemia, and sudden cardiac death, in addition to occlusive vascular disease. About half of cardiovascular deaths in both dialysis and transplant patients are caused by cardiac arrest and heart failure (ANZDATA 2009) for which the predominant pathogenetic mechanisms include hypertension, volume expansion, vascular calcification, and arrhythmia, rather than platelet aggregation and thrombosis. Therefore, while we find that antiplatelet agents prevent occlusive vascular events (myocardial infarction) in CKD as expected, they have a lower overall effect on nonthrombotic causes of death (both vascular and nonvascular). The results of the present review are consistent with the effects of antiplatelet agents in primary prevention of cardiovascular events, which reduce nonfatal myocardial infarction by 20% but do not prevent stroke or vascular mortality with similar effects in men and women (ATT 2009). Notably, in that review, the authors concluded that aspirin may be of uncertain net value, because reducing occlusive events may not be outweighed by risks of major bleeding.
A previous meta-analysis of medical adjuvant treatment to increase patency of arteriovenous fistulae and grafts included placebo-controlled studies of antiplatelet agents, low-dose warfarin, or fish oil was published in 2008 (Osborn 2008). In that systematic review, antiplatelet agents were considered separately in analyses that combined access types (graft or fistula) and analyses included a maximum of only three studies and 41 events. Analyses in that review may have been insufficient to provide reliable estimates of the benefits or toxicity of antiplatelet agents on vascular access outcomes. Our review also differs from a second recent review of antiplatelet agents for the prevention of arteriovenous fistula thrombosis of 10 studies (approximately 2000 participants), as we considered the outcomes of suitability for dialysis or access maturation, summarised study risks of bias, and explored sources of heterogeneity within treatment effects (Coleman 2010).
Implications for practice
Overall evidence ratings and recommendations for antiplatelet agents to prevent cardiovascular and dialysis access outcomes in people with CKD using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system for grading evidence are summarised ( Table 3) (GRADE 2011). This systematic review has shown that antiplatelet treatment in people with CKD for approximately 12 months reduces the risk of myocardial infarction, but not stroke, or all-cause or cardiovascular mortality. Antiplatelet agents given at the time of access surgery reduce thrombosis or failure of vascular access, but effects on suitability for dialysis and access maturation remain unclear. Major bleeding is increased by one third across all antiplatelet drugs and stages of CKD. The relative benefits of treatment in kidney transplant recipients and with primary prevention strategies in CKD are insufficient to inform practice.
Current clinical practice guidelines recommend antiplatelet treatment approaches similar to that of the general population in individuals with acute coronary syndromes (KDOQI 2005; Renal Association 2010) and CKD. The absolute effects of antiplatelet agents on the prevention of myocardial infarction and bleeding are summarised quantitatively in people with different absolute baseline risks for these outcomes based on risk estimates from the present systematic review ( Table 4). This table shows that while antiplatelet treatment in people at high risk of myocardial infarction (dialysis patients) is reasonable (preventing 32 myocardial infarctions in 1000 patients treated for one year versus incurring seven major bleeding events), the harms of antiplatelet treatment probably exceed the benefits for people with lower annual risks of vascular events (transplant recipients or stage 1 to 2 CKD; two myocardial infarctions prevented per 1000 patients treated for one year versus seven major bleeding events), assuming that myocardial infarction and major bleeding are equivalent outcomes. Insufficient information about the nature of major bleeding events incurred (particularly intracranial haemorrhage) in the available studies will make a decision about risk and benefit for intermediate-risk people incompletely informed.
Overall, antiplatelet agents given for four to six months after vascular access surgery will substantially reduce the risk of access failure and thrombosis (preventing 20 patients per 100 at risk experiencing access thrombosis or patency failure over six months of treatment at the risk of three to four experiencing a major bleed), although the effects of antiplatelet therapy of the important outcome of suitability for dialysis remain uncertain.
Implications for research
There are currently few data for antiplatelet therapy to prevent cardiovascular events in kidney transplant recipients with chronic or acute coronary artery disease. Specific head-to-head studies of antiplatelet regimens in individuals with CKD and acute coronary syndrome or undergoing percutaneous coronary intervention are required, particularly comparing newer thienopyridines (prasugrel or ticagrelor) versus clopidogrel. Studies should be designed to use standardised criteria to capture systematically all cardiovascular outcomes and major bleeding events in studies in which severe CKD is not an exclusion criterion. More information is required on the relative benefits of glycoprotein IIb/IIIa inhibitors in people with CKD and the effects of therapy on cardiovascular mortality and bleeding. The role of antiplatelet therapy as a primary prevention strategy to reduce cardiovascular and all-cause mortality in individuals with CKD without existing cardiovascular disease appears to be lower research priority.
- We wish to thank the referees for their advice and feedback during the preparation of this review.
- The authors would like to thank all study authors who responded to our queries about their studies. We received additional unpublished data from Drs James, Wiviott, Ferris, Lincoff, Balog, Wolski, Baigent, Kaufman, Topol, and Shao.
- The authors wish to thank Ms Narelle Willis, Managing Editor of the Cochrane Renal Group and Ms Ruth Mitchell, Trials Search Coordinator of the Cochrane Renal Group for their help with this review.
Data and analyses
- Top of page
- Authors' conclusions
- Data and analyses
- What's new
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
Appendix 1. Electronic search strategies
Appendix 2. Risk of bias assessment tool
Feedback concerning conclusions, 9 May 2013
Thank you for a much needed review addressing the gaps in literature regarding the risks and benefits of antiplatelets in the chronic kidney disease (CKD) population. We thought the literature search was very thorough and well done. However, we came up with a few questions upon reading this review and felt that the stated conclusion "antiplatelets reduce myocardial infarction...including those with early stages of CKD who do not have clinically-evident occlusive cardiovascular disease" may not be accurately reflected by the presented data.
Looking at the first primary outcome - fatal and non-fatal myocardial infarction (MI), it was unclear whether the population studied was addressing primary prevention, secondary prevention or acute treatment of MI as the included populations had different cardiovascular histories. Of the two studies that were given the most weight in the analysis (HOT Study 2010 and PURSUIT Study 1998), one investigated primary prevention of MI using ASA versus placebo, while the other investigated acute treatment of MI using eptifibatide + ASA + heparin compared to ASA + heparin. In the non-CKD population, efficacy of antiplatelets is dependent on the indication (ie. primary or secondary prophylaxis or treatment). Different antiplatelets also have different places in therapy.
We therefore feel that it may be inappropriate to pool these trials together as they were investigating different populations.
In this same analysis, there were also multiple interventions such as single antiplatelets versus placebo (HOT study 2010, Ell study 1982, Creek 1990, Dember 2008, STOP study 1995, UK-HARP-I study 2005, ETDRS 1992), dual antiplatelets versus placebo (Kaufman 2003), as well as dual antiplatelets versus single antiplatelet agents (CREDO study 2008, CHARISMA study 2009, EPILOG study 1997, EPIC study 1994, EPISTENT study 1998, Dixon study 2009, RAPPORT study 1998, PURSUIT study 1998, and IMPACT II 1997). With both placebo and antiplatelet in the "control" arms of one meta-analysis, comparison groups and treatment groups are not clearly delineated from one another. As this was unclear, readers may be misled into believing that the effect is driven purely from antiplatelet compared to placebo, when this is not the case. Even pooling the data on the seven placebo-controlled trials may be inappropriate as they were studied in different patient
populations and indications (e.g. primary prevention, non-cardiovascular outcomes). Similarly, the "treatment arms" of the meta-analysis contained one or more antiplatelet agents, which may have
biased the result towards the treatment arm over single agent or placebo "control". This can also make it difficult to isolate the beneficial agent in the dual antiplatelet studies. Due to the differences in treatment arms and patient populations, we feel it would valuable to investigate the outcomes of these factors in separate analyses.
It should also be mentioned that the patients included in this review were derived as subgroups from larger studies with different baseline cardiovascular risk factors (e.g. diabetes, coronary artery disease, hypertension, etc). As a result, one cannot conclude that patients with only CKD, and no additional cardiovascular risk factors, would benefit from antiplatelet use to decrease cardiovascular outcomes such as fatal and non-fatal MIs. Dixon 2009 and Dember 2008 were two
studies enrolling hemodialysis patients with a primary outcome of AV graft patency or thrombosis; fatal and non-fatal MIs were only reported as an adverse effect and could have been under-reported in the study.
We commend the authors for assessing bias in the included trials and for performing a sensitivity analysis to explore the impact of the bias. We feel that with the relatively high percentage of unclear or high risk of biases that exist in the trials, it would have been beneficial for the authors to report on the results of their sensitivity analyses to clarify the role of the bias and to substantiate the reported results.
We feel that the author's conclusion "antiplatelet agents reduce myocardial infarction" may be too broad of a conclusion to be drawn based on the analysis that was performed looking at fatal and non-fatal MI. As well, their specific reference to "patients with early stages of CKD who do not have a clinically-evident occlusive cardiovascular disease" suggests this effect is shown in the CKD
population when using antiplatelets for primary prevention; however, this aspect was not separated out in their analysis. We feel that the pooling of studies with varying patient populations and treatments is not appropriate in helping clinicians determine whether antiplatelets provide any benefit for MI in patients with CKD. While we did not explore the other identified primary outcomes in this review, we wonder if similar concerns exist for not only the efficacy but also the safety outcomes. We would appreciate an investigation into single antiplatelet therapy versus placebo for various cardiovascular indications. We hope the authors will provide clarification and address these concerns in their future updates.
We look forward to hearing your response to our comments.
Gloria Su, BSc. Pharm
Wan-Yun Polinna Tsai, BSc. Pharm
Megan Harbin, BSc. Pharm
Asal Taheri, BSc. Pharm
Aaron M Tejani, BSc. Pharm, PharmD
Thanks for the constructive comments.
1. Primary versus secondary prevention versus acute treatment
We combined treatment estimates for all available studies comparing antiplatelet therapy (with or without standard therapy) versus placebo/no treatment (with or without standard therapy alone) to examine treatment effects, which is a standard starting point for meta-analyses. For the outcome of fatal or nonfatal myocardial infarction, there was little or no heterogeneity in the treatment effects observed in all the available trials, suggesting that treatment estimates could be appropriately summarised into a single effect size.
While not necessary in the absence of significant heterogeneity, we explored for pre-specified trial-level variables that might have modified the treatment estimates that we observed. We specifically wished to know whether treatment effects differed for patients with existing cardiovascular disease compared to those without cardiovascular disease but this was not feasible due to as we found insufficient numbers of studies that were clearly primary prevention or secondary prevention studies. However, the lack of heterogeneity in the overall summary estimate suggests that antiplatelet agents have similar effects irrespective of the presence or absence of cardiovascular disease.
2. Multiple interventions:
Unlike the relative lack of primary versus secondary prevention trials, there was sufficient studies to explore any differences in treatment effects based on the class of antiplatelet used. While there were numerous different strategies for antiplatelet treatment in contributing trials, all the treatment interventions could be characterised by an antiplatelet agent in addition to standard care versus no treatment/placebo in addition to standard care. We have called this antiplatelet therapy versus control to acknowledge the heterogeneity of the intervention strategies used (rather than antiplatelet treatment versus placebo).
We used stratified analyses according to overall class of antiplatelet drug where possible but there was lack of power from available studies to understand fully all the various treatment effects for each individual antiplatelet regimen. An individual patient meta-analysis would be needed to give a more fine-grained understanding of the different interventions and their combinations in the CKD population.
3. Deriving patients from subgroups of larger studies:
Patients with CKD were evaluated in post-hoc analyses of larger trials in broader populations. These included trials in populations with acute coronary syndromes requiring percutaneous coronary artery procedures, patients with hypertension and those with diabetes mellitus. Trials of treatment tended to use different interventions (glycoprotein IIb/IIIa inhibitors with or without clopidogrel) whereas trials of primary or secondary prevention did not use these agents, preventing useful stratified analyses for either class of agent or cardiovascular prevention in these trials. We have concluded that the lack of a priori assessment of glycoprotein IIb/IIIa inhibitors in people with CKD is an important limitation of the current evidence.
4. Potential under-reporting of clinical outcomes
We agree that many trials were not designed to evaluate mortality and cardiovascular outcomes and that these events were reported in an ad hoc fashion (not prespecified) which may have underestimated their frequency. We include evaluation of this aspect of trials when considering whether they are at risk of bias due to selective reporting of expected outcomes.
5. Risks of bias
We did not specify risk of bias items as sources of heterogeneity we would explore in stratified analyses. In further updates of this review and if deemed appropriate and feasible, we will explore attrition bias and allocation concealment as potential sources of heterogeneity in subgroup or sensitivity analyses.
In conclusion, we thank Dr Su and others for constructive comments to this review. We agree that the review cannot provide high quality information about antiplatelet agents as primary prevention for cardiovascular disease in people with CKD. We acknowledge the limitations of studies in which adults with CKD were studied post hoc and which are heterogeneous for presence of cardiovascular disease and antiplatelet agent studied. We agree that clinical events may be under-reported in available studies and will explore in future versions of this review the effects of risk of bias on the estimated treatment effects of antiplatelet treatment in CKD.
Last assessed as up-to-date: 24 January 2011.
Protocol first published: Issue 11, 2010
Review first published: Issue 2, 2013
Contributions of authors
- Draft the protocol: MR, SP
- Study selection: MR, LDM, SP
- Extract data from studies: MR, LDM, SP
- Enter data into RevMan: MR, LDM, SP
- Carry out the analysis: MR, LDM, SP
- Interpret the analysis: MR, LDM, SP, JC, VP, SZ, AW, MJ, GFMS
- Draft the final review: MR, LDM, SP, JC, VP, SZ, AW, MJ, GFMS
- Disagreement resolution: GFMS
- Update the review: SP, GFMS
Declarations of interest
Vlado Perkovic is supported by a fellowship from the Heart Foundation of Australia and various grants from the Australian National Health and Medical Research Council. He has received speakers fees from Roche, Servier and Astra Zeneca, funding for a clinical trial from Baxter, and serves on Steering Committees for trials funded by Johnson and Johnson, Boehringer Ingelheim, Vitae and Abbott. His employer conducts clinical trials funded by Servier, Johnson and Johnson, Roche and Merck.
Sources of support
- No sources of support supplied
- Suetonia Palmer, New Zealand.Don and Lorraine Jacquot Fellowship; Amgen Dompe - Consorzio Mario Negri Sud Fellowship
Differences between protocol and review
We included studies of antiplatelet agents of fewer than two months follow-up if they provided outcome data for vascular access outcomes.
Medical Subject Headings (MeSH)
Cause of Death; Hemorrhage [chemically induced]; Myocardial Infarction [*prevention & control]; Platelet Aggregation Inhibitors [adverse effects; *therapeutic use]; Primary Prevention; Randomized Controlled Trials as Topic; Renal Insufficiency, Chronic [*complications; mortality]; Stroke [*prevention & control]
MeSH check words
* Indicates the major publication for the study