Abstract
Background and objectives: Patients with end stage renal disease (ESRD) are often prescribed antiplatelet medications. However, these patients are also at increased risk of bleeding compared with the general population, and an aim was made to quantify this risk with antiplatelet agents.
Design, setting, participants, & measurements: A systematic review of the literature (Medline, EMBASE, Cochrane CENTRAL and Google Scholar databases) was done to determine the bleeding risk in ESRD patients prescribed antiplatelet therapy. The secondary outcome was the effect on access thrombosis. All case series, cohort studies and clinical trials were considered if they included ten or more ESRD patients, assessed bleeding risk with antiplatelet agents, and lasted for more than 3 mo.
Results: Sixteen studies, including 40,676 patients, were identified that met predefined inclusion criteria. Due to study heterogeneity and weaknesses in methodology, bleeding rates were not pooled across studies. However, the bleeding risk appears to be increased for hemodialysis patients treated with combination antiplatelet therapy. The results are mixed for studies using a single antiplatelet agent. Antiplatelet agents appear to be effective in preventing shunt and central venous catheter thrombosis, but not for preventing thrombosis of arteriovenous grafts.
Conclusion: The risks and benefits of antiplatelet agents in ESRD patients remain poorly defined. Until a clinical trial addresses this in the dialysis population, individual risk stratification taking into account the increased risk of bleeding should be considered before initiating antiplatelet agents, especially in combination therapy.
Chronic hemodialysis (HD) has been feasible for almost half a century now, ever since the introduction of the Quinton-Scribner arteriovenous shunt in 1960 (1). The direct access to the circulation afforded initially by the shunt and subsequently by the arteriovenous graft (AVG), arteriovenous fistula (AVF) and central venous catheter (CVC) remains a critical component of the HD procedure. Yet, from the earliest days, the vascular access has been dogged by problems related to clotting. This has taken the form of neo-intimal hyperplasia and stenosis leading to occlusion for AVGs, high primary failure rates for AVFs and malfunction with the need for frequent replacements for CVCs. Indeed, vascular access dysfunction represents a significant proportion of the morbidity affecting hemodialysis patients with its attendant health care related expenditure (2). Antiplatelet agents represent an obvious strategy to tackle this issue.
However, for more than a century, patients with chronic kidney disease (CKD) have been recognized to be at a higher risk of bleeding resulting from impaired platelet function, as well as abnormal platelet-vessel wall interactions (3–5). Although most commonly manifested clinically as minor bleeding from skin and mucus membranes, it can frequently occur as gastrointestinal or cerebral hemorrhage, and bleeding at other systemic sites as well (6–9). Aspirin has been shown to worsen the platelet dysfunction in CKD patients in vitro (10). Epidemiologic studies in the general population have revealed a doubling of the risk of gastrointestinal bleeding even with low-dose (< 300 mg daily) aspirin use (11). However, a previous meta-analysis could not reliably estimate the risk of bleeding in HD patients because of the small number of events (46 total extracranial bleeds) (12).
Studies of antiplatelet agents in access preservation for HD patients have had mixed results, with a randomized controlled trial showing lack of efficacy of clopidogrel (13) and an observational study suggesting otherwise (14). Given the potential risk of antiplatelet agents, we performed a systematic review to evaluate the bleeding risk associated with long-term antiplatelet use in the end stage renal disease (ESRD) population. Secondarily, we assessed the efficacy of antiplatelet agents in reducing access thrombosis. We hypothesized that the use of antiplatelet agents, compared with control groups or placebo, would be associated with an enhanced bleeding risk.
Materials and Methods
A comprehensive and open literature search was performed using the following databases: MEDLINE, 1966 to week 2 June 2008; EMBASE, 1980 to week 2 June 2008, Cochrane Central Register of Clinical Trials (1st quarter 2008) and Google Scholar (until 2008). The search strategy was devised to capture clinical trials as well as observational studies of antiplatelet agents in the dialysis population (search strategy available on request). Abstracts and citation titles were independently reviewed by two investigators (SH, DZ) and citations considered relevant by either reviewer were retrieved to be evaluated as full texts. All these potentially relevant articles were again reviewed by two investigators (SH, DZ) to be considered for inclusion. Disagreements on inclusion were resolved by consensus discussion and arbitration by a third author (RH).
All case series, cohort studies and randomized controlled trials were considered if they included ten or more ESRD patients. The studies also had to have assessed bleeding risk with antiplatelet agents. We only considered studies that had three or more months of follow-up to ensure that bleeding rates were related to antiplatelet exposure. Unpublished abstracts, review articles and editorials were excluded. The reference lists of all eligible studies were also reviewed for potentially relevant studies.
Two investigators (SH, DZ) independently abstracted data from all eligible studies using a standardized form. The methodological data collected included study design, participant number, description of drop-outs and loss to follow-up, study duration and study quality. Study quality for randomized clinical trials was assessed using the Jadad scale, which measures blinding, randomization, withdrawals and drop-outs (16). Study quality for observational studies was assessed using the Newcastle-Ottawa scale, which measures selection, comparability and assessment of outcomes (43). The clinical data abstracted included the antiplatelet intervention, a priori bleeding risk definition, and bleeding and thrombosis rates. Definitions of bleeding events provided by the study authors were used in the synthesis. In the event that bleeding rates could not be calculated from the data provided, we attempted to contact the investigators of the studies to provide the missing information. Differences on data abstracted were resolved by consensus and discussion with another investigator (DF).
Results
Literature Search
The initial search of the available literature yielded 957 citations, of which 69 were assessed in greater detail. Fifty-three of the 69 studies were not included for the following reasons: < 3 mo duration (n = 22), addressed a different question (n = 14), <10 patients (n = 7), abstract form or were publications of the protocol only (3 + 2), duplicate studies (n = 2), internal company reports (n = 2) and nonseparation of hemodialysis patients from CKD (n = 1). A total of 16 studies (17–32) was included in the systematic review (Figure 1). For two studies included, separate data on bleeding events in dialysis patients, which was not reported in the published data, was obtained from correspondence with the authors (30,32).
Figure 1.
Flow diagram of search results and selection of studies.
Study Design Characteristics
Table 1 lists references, study design, sample size and follow-up. There were nine randomized controlled trials (RCTs), four prospective cohort and three retrospective cohort studies published between 1975 and 2008. Two of the RCTs had a crossover design and one of them (Kooistra) had a one-week washout period. Although the other study (Kaegi) did not include a washout phase, the bleeding events occurred only in the initial phase, before crossover. A total of 40,676 patients with ESRD were included in these studies with 9186 patients being exposed to an antiplatelet agent(s). However, most of the studies were small, with a median sample size of 113 (16 to 28,320), and only two studies included more than 2500 patients (the number required to reliably detect an adverse event rate of 5 ± 2%). Aspirin was the most common antiplatelet agent used as a sole agent in seven studies and in combination with warfarin or other antiplatelet agents in four studies. The other antiplatelet agents studied were ticlopidine, clopidogrel, sulfinpyrazone and dipyridamole. The median follow-up was 8 mo (range 3 mo to 6 yr) and the indication for antiplatelet therapy was for the prevention of vascular access thrombosis in all but four studies. The studies were heterogeneous with respect to the type of vascular access, type and dose of antiplatelet agent used, length of follow-up and definition of bleeding risk. Given the heterogeneity among studies and weaknesses in study methodology, we could not pool bleeding rates across studies. Instead, we qualitatively reviewed the pertinent results of each study. The Jadad score ranged from 1 to 5 (maximum score for the scale = 5) for the RCTs and the Newcastle-Ottawa scale ranged from 5 to 7 (maximum 9 “stars”).
Table 1.
Included studies of antiplatelet use in HD patients: Study design and treatments
| Author | Trial Type | Control Group | N | Groups | Drop Out; Loss to Follow-Up | Number Analyzed | Study Length (months) | Jadad Score (max = 5) | N-O Scale (max = 9) |
|---|---|---|---|---|---|---|---|---|---|
| Kaegi (1975) | RCT, crossover | Placebo | 62 | S, Pa | 17 | 45 | 6 | 4 | – |
| Kobayashi (1979) | RCT | No | 119 | T:47b | 4 | 47 | 3 - 8 | 2 | – |
| Harter (1979) | RCT | Placebo | 44 | A:19 P:25 | NR | 44 | 5 | 4 | – |
| Kobayashi (1980) | RCT | Placebo | 107 | T:50 P:57 | 7 | 100 | 3 | 4 | – |
| Domoto (1991) | Prospective cohort | None | 16 | A + S:16 | 1 drop out | 15 | 12 | – | 5 |
| Kooistra (1994) | RCT, crossover | Placebo | 153 | A, Pa | 35 | 137 | 3 | 4 | – |
| Sreedhara (1994) | RCT | Placebo | 107 | D:29 A:26 D + A:28 P:24 | 11 drop outs; 2 lost to FU | D: 24 A: 25 D + A: 24 P: 23 | 72 | 4 | – |
| Kaufman (2003) | RCT | Double placebo | 200 | A + Cl:104 P:96 | A + Cl:30 P:36 | 200 | ∼7 | 5 | – |
| Obialo (2003) | Prospective cohort | Yes | 63 | A:21, W:11 C:31 | NR | 63 | ∼4 | – | 6 |
| Wasse (2003) | Retrospective | Yes | 10,842 | A:1527 C:9315c | NR | 10842 | ∼32 | – | 7 |
| Trimarchi (2005) | Prospective Non-randomized | Yes | 19 | Cl:8 C:11 | NR | 19 | ∼12 | – | 5 |
| Coli (2006) | RCT | No | 144 | W + T: 81 dW + T:63 | 11 deaths | 144 | 12 | 1 | – |
| AbdulRahman (2007) | RCT | Yes | 58 | W:20, A:19 C:19 | NR | 58 | 12 | 1 | – |
| Ethier (2007) | Prospective cohort | Yes | 28,320 | A:6,573 C:21,707 | NR | 28320 | ? | – | 7 |
| Holden (2008) | Retrospective | Yes | 255 | W: 89, A:107 A + W:50 C: 178 | 0 | 255 | 1028 person- years | – | 7 |
| To (2008) | Retrospective | Yes | 167 | W:11, A72, C 84 | 4 recovered, 8 lost to FU | 155 | 26 | 7 |
Cross-Over Studies.
Only patients that received medication for 12 wk or more are included.
Many subgroups considered including 280 patients on antiplatelet agents but not divided by type; not included in this study.
Patients randomized to warfarin + ticlopidine 12 h post-operatively versus or after the first thrombotic event.
A, aspirin; Cl, clopidogrel; C, control; D, dipyridamole; P, placebo; S, sufinpyrazone; T, ticlopidine; W, warfarin; NR, not reported; N-O scale, Newcastle-Ottawa scale.
Antiplatelet Agents and Bleeding Risk
In Table 2, the type of antiplatelet agent, definition of bleeding complications and event rates are shown. One of the 16 studies (6%) used an adequate a priori definition of hemorrhage to report the different types of bleeding (25). The definition of bleeding in most other studies was either not reported at all (ten studies) or was inadequate (five studies; e.g., gastrointestinal bleeds only; undefined “major bleeding episodes”).
Table 2.
Intervention and bleeding events
| Study | Intervention | Bleeding |
Bleeding Events | Reported Change in Bleeding with antiplatelet agents | |
|---|---|---|---|---|---|
| Defined a priori | Definition used | ||||
| Sulfinpyrazone | |||||
| Kaegi (1975) | Sulfinpyrazone 200 mg tid, Placebo | No | GI bleeding | S: 2 of 45 P: 1 of 45 | Unchanged |
| Aspirin | |||||
| Harter (1979) | Aspirin 160 mg OD, Placebo | No | Transfusions | A: 5 (4.8units/patient) P:13 (3.9units/patient) | Unchanged |
| Kooistra (1994) | Aspirin 30 mg OD, Placebo | No | Bleeding incidents not due to HD | A: 2 P: 5 | Decreased |
| Wasse (2003) | Aspirin, Control | Yes | Upper GI bleed defined by billing codes | A:24.1/1000 P-Y C:22.6/1000 P-Y | Unchanged |
| Obialo (2003) | Aspirin 325 mg OD or Control (no A or W) [or Warfarin INR 2–3] | No | GI bleeds | A: 5 (24%) C: 0 (0%) (W: 2 [18%]) | Increased |
| Abdul-Rahman (2007) | Aspirin 81 mg OD or Control or [Warfarin INR 1.5–2.0] | Yes | Major bleeding episodes retroperitoneal, intra-articular or cerebral or 2g/dl decrease in Hgb | A:0 C:0 [W:0] | Unchanged |
| Ethier (2007) | Aspirin, Control | Yes | GI bleeding, Subdural Hematoma | A: 4.29 events/100pt yrs C: 3.74 events/100pt yrs A 0.17 events/100pt yrs vs 0.24 events/100 pt yrs | Unchanged Decreased |
| Holden (2008) | Aspirin or Control or [Aspirin + Warfarin or Warfarin] | Yes | Major Bleed Hospitalized, CNS, retro-peritoneum, 2 units RBC | A 4.4% (2.3–7.7%)/P-Y C: 0.8% (0.2–2.0%)/P-Y (A + W 6.3% [2.1–14.8%]/P-Y W:3.1% [1–7.3%]/P-Y) | Increased |
| Ticlopidine | |||||
| Kobayashi (1979) | Ticlopidine 100, 200, 300 mg | No | Bleeding Tendency | 4 of 47 (epistaxsis, bleeding with shaving cut) | No control group |
| Kobayashi (1980) | Ticlopidine 100 mg bid, Placebo | No | Major or serious bleeding | T: 4 (2 gingival, 2 serious) P: 3 (1 gingival, 2 serious) | Unchanged |
| Clopidogrel | |||||
| Trimarchi (2005) | Clopidogrel 75 mg OD, Control | No | “Major bleeding episodes” | Cl: 0 C: 0 | Unchanged |
| Combinations: Antiplatelet + Warfarin or Antiplatelet Compared to Warfarin | |||||
| Coli (2006) | Ticoplidine 250 mg OD + Warfarin INR 1.8–2.5 | No | Bleeding events | T + W: 0; T + W (post 1st thrombosis): 0 | No control group |
| To (2008) | Warfarin or ASA | Yes | CNS bleed, requiring hospitalization or blood transfusion | W: 3 of 11 (27%) vs 28 of 144 (19%) A: 13 of 72 (18%) vs 18 of 83 (22%)a | Unchanged |
| Combinations: 2 Antiplatelet agents | |||||
| Domoto (1991) | Aspirin 85 mg od + Sulfinpyrazone 200 mg tid | No | No specific definition | A + S: 6-GI bleeds; 2-Access bleeds; 1- Retro- peritoneal bleed (9/15) | No control group |
| Sreedhara (1994) | Dipyramidole 75 mg tid, Aspirin 325 mg daily, Both (A + D), Placebo | No | GI adverse events | D: 5/29 A: 3/26 A + D: 5/28 P: 2/24 | Unchanged |
| Kaufman (2003) | Aspirin 325 mg + Clopidogrel 75 mg daily, Placebo | Yes | As major, intermediate and minor | A + C: 44/104 (67 bleeding events in 44 participants) P: 23/96 (38 bleeding events in 23 participants) | Increased |
OD, once daily; bid, twice daily; tid, thrice daily; GI, gastrointestinal; P-Y, patient years; A, aspirin; C, control; Cl, clopidogrel; P, placebo; D, dipyramidole; S, sulfinpyrazone; T, ticlopidine.
1 patient on W + A.
When aspirin was studied in combination with clopidogrel, the bleeding risk was found to be doubled (HR 1.98; P = 0.007) and the study was terminated early. The absolute risk increase for a first bleeding event in the aspirin/clopidogrel-treated group was 0.18 (CI 0.06 to 0.31), and the number needed to harm was only 5.6. Furthermore, in another antiplatelet combination study, nine of 16 (56%) patients who were on aspirin and sulfinpyrazone had a bleeding event including one retroperitoneal bleed (22). One of the 3 studies using combinations of antiplatelet agents demonstrated that bleeding risks were not increased (24). However, in this study, the definition of bleeding was not described a priori and only reported as gastrointestinal adverse events.
Of the seven studies where aspirin was used as the sole antiplatelet agent, three reported an increased risk of bleeding. Among the studies where a fixed dose of aspirin was used, there appears to be dose-related increased risk of bleeding. The two studies where low dose was used (30 mg and 81 mg) did not report an increase in bleeding as compared with the study where 325 mg aspirin was used.
The bleeding risk with sulfinpyrazone (18), ticlopidine (19,21) and clopidogrel (28) does not appear to be increased when used in patients on hemodialysis compared with patients treated with placebo or control, even when used in combination with warfarin (29).
Antiplatelet Agents and Access Thrombosis Risk
In Table 3, the indication for antiplatelet agent, type of antiplatelet agent and access thrombosis events are shown. In arteriovenous shunts, the use of antiplatelet agents was associated with a 1.25 to threefold decrease in the risk of thrombosis. The effect was seen regardless of type of antiplatelet agent used. This calculation does not include the first Kobayashi et al. manuscript as the effect of ticlopidine on shunt thrombosis was graded as decreased, no change or increased. In all cases in that study, shunt thrombosis was either unchanged (14 of 47) or decreased (23 of 47) (7). Additionally, in that study we excluded the data from patients who had been followed for less than 12 wk (n = 72).
Table 3.
Intervention and thrombosis events
| Study | Intervention | Outcome | Number of events | Reported event rates and/or measures of association |
|---|---|---|---|---|
| Study Indication: Preservation of AV shunts | ||||
| Kaegi (1975) | Sulfinpyrazone 200 mg tid, Placebo | Shunt thrombosis rate per month | S: 58 of 45 patients P: 174 of 45 patients | 0.21 thrombi/month vs 0.64 thrombi/month (P < 0.001) |
| Kobayashi (1979) | Ticlopidine 100, 200, 300 mg | Shunt clot formation | T: Decreased in 23 of 47 | |
| Harter (1979) | Aspirin 160 mg daily, Placebo | Shunt thrombosis | A: 14 thrombi/6 patients (6 of 19 patients) P: 53 of 18 patients (18 of 25 patients) | 0.16 thrombi/month vs 0.46 thrombi/month (P < 0.005) |
| Kobayashi (1980) | Ticlopidine 100 mg bid, Placebo | Clot removal, surgery | Efficacy: T:34 of 47 improved P 15 of 53 improved | 0.84 thrombi/month vs 1.06 thrombi/month (P < 0.05) |
| Study Indication: Preservation of AV grafts | ||||
| Domoto (1991) | Aspirin 85 mg daily + Sulfinpyrazone 200 mg tid | Access thrombosis rate per month | A + S: 4 of 15 patients thrombosis (total # thrombosis = 4) C: 15 of 15 patients thrombosis (total # thrombosis = 70) | 0.04 thrombi/month vs 0.12thrombi/month (P < 0.001) |
| Sreedhara (1994) | Dipyramidole 75 mg tid Aspirin 325 mg daily Both (D + A) Placebo | Thrombosis | D: 9 of 29 (Type 1 4 of 23, Type 2 5 of 6)a | RR 0.35 (CI 0.15–0.8, P = 0.02) |
| A: 13 of 26 (Type 1 10 of 20, Type 2 3 of 6) | RR 1.99 (CI 0.88–4.48, P = 0.18) | |||
| B: 11/28 (Type 1 5 of 22, Type 2 6 of 6) | ||||
| P: 10 of 24 (Type 1 6 of 19, Type 2 4 of 5) | ||||
| Kaufman (2003) | Aspirin 325 mg daily + Clopidogrel 75 mg daily, Placebo | Thrombosis hazard rates | Annual Hazard Rate A + Cl: 0.47 (CI 0.31–0.71); P: 0.59 (CI 0.39–0.87) | HR 0.81 (CI 0.47–1.40, P = 0.45) |
| Trimarchi (2005) | Clopidogrel 75 mg daily, Control | Graft Thrombosis | Cl 0 of 8 patients C: 10 of 11 patients | Graft Patency 350.8 (166) days vs 86.8 (69), p=<0.001 |
| Study Indication: Prolong catheter patency | ||||
| Obialo (2003) | Aspirin 325 mg/d, Warfarin INR 2–3, Control | Catheter patency duration (days)–includes removal for bacteremia | W: 111 ± 17 (mean duration catheter patency) | Patency at 120 days: W 73% |
| A: 114 ± 18 | A 91% | |||
| C: 68 ± 37 | C 29% | |||
| Coli (2006) | Ticoplidine 250 mg/day + Warfarin INR 1.8–2.5 | Catheter thrombosis or malfunction | T + W: 10 of 81 patients T + W (started post thrombotic event): 33 of 63 patients | 0.16 thrombosis/yr vs 1.65 thrombosis/yr (P < 0.001) |
| Abdul-Rahman | Warfarin (1.5–2.0) ASA 81 mg OD, Control | Catheter thrombosis | W:4 of 20 patients A:4 of 19 patients C:9 of 19 patients | Malfunction free catheter survival at 12 months W: 75%, A:68.4%, C:36.8% |
Includes patients that did not complete the protocol–presumes all did not have thrombosis.
A, aspirin; Cl, clopidogrel; C, control; D, dipyridamole; P, placebo; S, sufinpyrazone; T, ticlopidine; W, warfarin; RR, relative risk; HR, hazard ratio.
In arteriovenous grafts (AVGs), the results are mixed. The study with the most rigorous methodology included two antiplatelet agents versus placebo and was stopped early secondary to an enhanced bleeding risk without a statistically significant benefit in reducing access thrombosis (25). In the Sreedhara et al. study, aspirin alone appeared to increase the risk of thrombosis (nonsignificant); dipyridamole alone or in combination with aspirin was associated with a decreased risk of thrombosis (24). However, in the calculation of thrombosis risk, 11% of the patients did not complete the study but are presumed to not have clotted their accesses. This had the greatest effect in the dipyridamole and combination groups, in which five of 23 and four of 22 had incomplete data, respectively. Importantly, in both of these studies, antiplatelet agents appeared to be more efficacious if they were used as prophylaxis (before any thrombotic event) as opposed to treatment after a thrombotic episode (24,25). Although the risk of thrombosis was reduced in the Domoto et al. prospective cohort study of bovine artery grafts with aspirin and sulfinpyrazone, the bleeding risk was also high. Lastly, in the Trimarchi study, the AVG patency was significantly improved with clopidogrel (350 [±166] compared with 86.8 [±69] days; P = 0.00); however, patient selection bias may have affected the results, as only two of the patients on clopidogrel died compared with all the patients that were not treated with an antiplatelet agent. Additionally, there are discrepancies in the numbers of patients who developed access thrombosis in different sections of the manuscript.
There were three studies addressing antiplatelet agents in the prevention of central venous catheter thrombosis (26,29,30). In the first study, patients treated with warfarin or aspirin were compared with patients that were not prescribed either of these medications. Both warfarin and aspirin improved access patency from an average of 68 d to approximately 111 d (26). Similarly, in a randomized controlled trial in which 20 patients received warfarin, 19 patients received low-dose aspirin and 19 patients served as controls, one year access survival was about two times greater for the treatment groups compared with the control group (30). However, although the study is reported to be double blinded, there is no mention of a matching placebo in the two treatment groups or a double placebo in the control group. Since the primary outcome includes inability to maintain blood flows of >250 ml/min during dialysis, which may be operator dependent, the trial results should not be considered definitive. The other study was an RCT in which patients received ticlopidine and warfarin either within 24 h post central venous catheter insertion or delayed until a first thrombotic event (29). The early initiation of ticlopidine and warfarin was associated with a reduced risk of access thrombosis from 1.65 access thrombosis/yr to 0.16 access thrombosis per year. However, the two groups of patients are not comparable, as the second group of patients had to have a thrombotic event before being enrolled in the trial, a known risk factor for access failure.
Discussion
We sought to determine if antiplatelet agents increase the risk of bleeding in stage 5 CKD patients and secondarily to determine if antiplatelet agents reduce the risk of access thrombosis.
The bleeding risk for patients on dialysis treated with antiplatelet agents appears to be related to the prescribed number of antiplatelet agents and the type of antiplatelet agent. In two of the three studies with two antiplatelet agents, the bleeding risk was increased. In the Kaufman study, which had the best scientific rigor, the risk of a major or a minor bleed for patients on hemodialysis was estimated to be two times and five times greater than patients that had been enrolled in the Clopidogrel in Unstable Angina to Prevent Recurrent Events Trial (CURE) despite the use of the same two antiplatelet agents (33). Importantly, the combination of these two antiplatelet agents has become the standard of care for patients undergoing percutaneous coronary intervention and with acute coronary syndromes partially due to the results of the CURE study (34,35). A trend toward increased risk of bleeding with the use of these two agents in combination (especially with aspirin doses >100 mg) has been recently reported even in the nondialysis population (44). In HD patients, this bleeding risk is likely to be significantly higher.
The studies in which aspirin was used singly have reported conflicting results. There are a number of potential explanations for the differences in outcomes. Variation in the definitions of bleeding risk, selection bias (low-risk patients receiving aspirin), confounding by indication and co-interventions (warfarin, proton pump inhibitors) are some of them. There might be a dose- related effect of aspirin on bleeding risk (44). Additionally, in the Ethier et al. study, the relative risk of a subdural hematoma was 0.56 (CI 0.30 to 1.07) in favor of aspirin. While aspirin may be protective, the more likely explanation is confounding by indication and co-interventions.
Lastly, the bleeding risk may be different for different antiplatelet agents. Patients treated with sulfinpyrazone, clopidogrel and ticlopidine do not appear to have a significantly increased bleeding risk compared with placebo or control patients. In the Holden et al. retrospective cohort study, 34 patients had been treated with clopidogrel for a total of 46 person-years of exposure without a bleeding event. Due to the limited exposure, this medication was not included in the analysis of bleeding risk. However, the low risk of bleeding with clopidogrel is supported by the Dialysis Access Consortium (DAC) study in which patients with a newly created arteriovenous access randomized to clopidogrel versus placebo for 6 wk did not show a difference in bleeding risk (13).
The overall quality of many of the included studies, the small numbers of patients, the lack of predefined bleeding definitions and different anti-platelet dosing/indications in individual studies limit the ability to quantitate the overall bleeding risk in HD patients on antiplatelet medications.
The literature is also mixed with respect to the benefits of antiplatelet agents in reducing access thrombosis. Antiplatelet agents do appear to be effective in the reduction of shunt thrombosis. This is only of historical interest, as this type of access is no longer used for hemodialysis. However, the decreased risk of shunt thrombosis might be expected to translate into a decreased risk of CVC thrombosis as the synthetic materials are similar. We identified three studies in which an antiplatelet agent used alone (aspirin) or in combination with warfarin (ticlopidine) did appear to decrease the risk of CVC thrombosis. Again, study quality limits definitive conclusions.
For AVGs, aspirin alone actually appeared to increase the risk of AVG clotting. It is unclear if this may have contributed to the lack of benefit in the study in which aspirin and clopidogrel were used in combination. Importantly, in both of these studies, the outcomes for AVG thrombosis appear to be different if the medication is used for prophylaxis (helpful) or treatment (not helpful). This information may be useful in future randomized controlled trials. No studies satisfying our inclusion criteria were identified for the preservation of AVF. However, the use of clopidogrel for 6 wk has since been shown to be ineffective in improving the primary function of AVFs despite a decreased risk of thrombosis (13). Additionally, aspirin did not affect primary or secondary fistula patency using the DOPPS database (36).
This analysis cannot answer the issue of the use of antiplatelet agents in prevention and/or treatment of cardiovascular disease in dialysis patients. Two of the studies included in the assessment of bleeding did examine the ability of aspirin to reduce the risk of vascular events; overall, both were negative (22,30). These findings are consistent with two other studies in which subgroup analysis was performed for CKD patients (37,38). A recent retrospective study suggests that, in fact, antiplatelet agents might increase mortality in hemodialysis patients (45), although there is likely residual confounding by indication, which might have biased these results. Overall, given the limitations of these analyses, we believe clinical equipoise exists in this regard until an appropriately designed RCT is done in this subpopulation.
Dialysis patients are at increased risk of bleeding compared with the general population (39,40,41). Even in the erythropoietin era, the transfusion rate is 0.66 transfusions per patient per year (42). The use of a single antiplatelet agent does not appear to increase the risk of bleeding, with the possible exception of aspirin in which the study results are mixed. However, antiplatelet agents used in combination do appear to significantly increase the risk of bleeding, which should be considered when initiating such a therapeutic strategy. Importantly, the current studies do not support the use of antiplatelet agents for maintenance of the hemodialysis vascular access with the exception of CVCs in which antiplatelet agents may have a role. Until rigorously done RCTs are able to properly address the risk-benefit question, individualized risk assessment should be considered before initiating antiplatelet agents in this population.
Disclosures
None.
Acknowledgments
The authors would like to thank Drs. To and Ethier for providing additional information from their studies. An earlier version of this study appeared in abstract form and was presented at the American Society of Nephrology meeting, October 31 through November 2, 2007 (SA-PO790).
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
References
- 1.Quinton W, Dullard D, Scribner BH: Cannulation of blood vessels for prolonged hemodialysis. Trans Am Soc Artif Intern Organs 6: 104–113, 1960 [PubMed] [Google Scholar]
- 2.Feldman HI, Kobrin S, Wasserstein A: Hemodialysis vascular access morbidity. J Am Soc Nephrol 7(4): 523–535, 1996 [DOI] [PubMed] [Google Scholar]
- 3.Reisman D: Hemorrhages in the course of Bright's disease, with especial reference to the occurrence of a hemorrhage diathesis of nephritic origin. Am J Med Sci 134: 709–712, 1907 [Google Scholar]
- 4.Kaw D, Malhotra D: Platelet dysfunction and end-stage renal disease. Semin Dial 19: 317–322, 2006 [DOI] [PubMed] [Google Scholar]
- 5.Castillo R, Lozano T, Escolar G: Defective platelet adhesion on vessel subendothelium in uremic patients. Blood 68: 337–342, 1986 [PubMed] [Google Scholar]
- 6.P.K.B. Jaminidar In: AR FRGD Nissenson, Editor, Gastrointestinal disease in patients with chronic renal failure, chapter 26 in clinical dialysis vol. 3, East Norwalk, Appleton & Lange, 1995, pp. 607–617 [Google Scholar]
- 7.Iseki K, Kinjo K, Kimura Y, Osawa A, Fukiyama K: Evidence for high risk of cerebral haemorrhage in chronic dialysis patients. Kidney Int 44: 1086–1090, 1993 [DOI] [PubMed] [Google Scholar]
- 8.Milutinovich J, Follette WC, Scribner BH: Spontaneous retroperitoneal bleeding in patients on chronic hemodialysis. Ann Intern Med 86: 189–192, 1977 [DOI] [PubMed] [Google Scholar]
- 9.Smetana SS, David E, Pelet D, Bar-Khayim Y: Subcapsular liver hematoma in a patient on chronic hemodialysis. Nephron 45: 323–324, 1987 [DOI] [PubMed] [Google Scholar]
- 10.Livio M, Benigni A, Vigano G, Mecca G, Remuzzi G: Moderate doses of aspirin and risk of bleeding in renal failure. Lancet 1: 414–416, 1986 [DOI] [PubMed] [Google Scholar]
- 11.Garcia Rodriguez LA, Hernandez-Diaz S, De Abajo FJ: Association between aspirin and upper gastrointestinal bleeding complications: Systematic review of epidemiological studies. Br J Clin Pharmacol 52: 563–571, 2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Antithrombotic Trialists' Collaboration Collaborative meta-analysis of randomized trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 324: 71–86, 2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Dember LM, Beck GJ, Allon M, Delmez JA, Dixon BS, Greenberg A, Himmelfarb J, Vazquez MA, Gassman JJ, Greene T, Radeva MK, Braden GL, Ikizler TA, Rocco MV, Davidson IJ, Kaufman JS, Meyers CM, Kusek JW, Feldman HI; Dialysis Access Consortium Study Group: Effect of clopidogrel on early failure of arteriovenous fistulas for hemodialysis: A randomized controlled trial. JAMA 299(18): 2164–2171, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hasegawa T, Elder SJ, Bragg-Gresham JL, Pisoni RL, Yamazaki S, Akizawa T, Jadoul M, Hugh RC, Port FK, Fukuhara S: Consistent aspirin use associated with improved arteriovenous fistula survival among incident hemodialysis patients in the dialysis outcomes and practice patterns study. Clin J Am Soc Nephrol 3(5): 1373–1378, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Elliott MJ, Zimmerman D, Holder RM: Warfarin anticoagulation in hemodialysis patients: A systematic review of bleeding rates. Am J Kidney Dis 50: 433–440, 2007 [DOI] [PubMed] [Google Scholar]
- 16.Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, McQuay HJ: Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Control Clin Trials 17: 1–12, 1996 [DOI] [PubMed] [Google Scholar]
- 17.Kaegi A, Pineo GF, Shimizu A, Trivedi H, Hirsh J, Gent M: The role of sulfinpyrazone in the prevention of arterio-venous shunt thrombosis. Circulation 52: 497–499, 1975 [DOI] [PubMed] [Google Scholar]
- 18.Kobayashi K, Maeda K, Koshikawa S: Antithrombotic therapy with ticlopidine in chronic renal failure patients on maintenance hemodialysis. J New Remedies & Clinics 28: 1032–1049, 1979 [DOI] [PubMed] [Google Scholar]
- 19.Harter HR, Burch JW, Majerus PW, Stanford N, Delmez JA, Anderson CB, Weerts CA: Prevention of thrombosis in patients on hemodialysis by low-dose aspirin. N Engl J Med 301: 577–579, 1979 [DOI] [PubMed] [Google Scholar]
- 20.Kobayashi K, Maeda K, Koshikawa S, Kawaguchi Y, Shimizu N, Naito C: Antithrombotic therapy with ticlopidine in chronic renal failure patients on maintenance hemodialysis-A multicenter collaborative double blind study. Thrombosis Research 20: 255–261, 1980 [DOI] [PubMed] [Google Scholar]
- 21.Domoto DT, Bauman JE, Joist JH: Combined aspirin and sulfinpyrazone in the prevention of recurrent hemodialysis vascular access thrombosis. Thrombosis Research 62: 737–743, 1991 [DOI] [PubMed] [Google Scholar]
- 22.Kooistra MP, van Es A, Marx JJ, Hertsig ML, Struyvenberg A: Low-dose aspirin does not prevent thrombovascular accidents in low-risk hemodialysis patients during treatment with erythropoietin. Nephrol Dial Transplant 9: 1115–1120, 1994 [DOI] [PubMed] [Google Scholar]
- 23.Sreedhara R, Himmelfarb J, Lazarus JM, Hakim RM: Anti-platelet therapy in graft thrombosis: Results of a prospective, randomized, double-blind study. Kidney Int 45: 1477–1483, 1994 [DOI] [PubMed] [Google Scholar]
- 24.Kaufman JS, O'Connor TZ, Zhang JH, Cronin RE, Fiore LD, Ganz MB, Goldfarb DS, Peduzzi PN; Veterans Affairs Cooperative Study Group on Hemodialysis Access Graft Thrombosis: For the veterans affairs cooperative study group on hemodialysis access graft thrombosis. Randomized controlled trial of clopidogrel plus aspirin to prevent hemodialysis access graft thrombosis. J Am Soc Nephrol 14: 2313–2321, 2003 [DOI] [PubMed] [Google Scholar]
- 25.Obialo CI, Conner AC, Lebon LF: Maintaining patency of tunneled hemodialysis catheters. Scand J Urol Nephrol 37: 172–176, 2003 [DOI] [PubMed] [Google Scholar]
- 26.Wasse H, Gillen DL, Ball AM, Kestenbaum BR, Seliger SL, Sherrard D, Stehman-Breen CO: Risk factors for upper gastrointestinal bleeding among end-stage renal disease patients. Kidney Int 64: 1455–1461, 2003 [DOI] [PubMed] [Google Scholar]
- 27.Trimarchi H, Young P, Forrester M, Schropp J, Pereyra H, Freixas E: Clopidogrel diminishes hemodialysis access graft thrombosis. J Vascular Access 6: 29–33, 2005 [DOI] [PubMed] [Google Scholar]
- 28.Coli L, Donati G, Cianciolo G, Raimondi C, Comai G, Panicali L, Nastasi V, Cannarile DC, Gozzetti F, Piccari M, Stefoni S: Anticoagulation therapy for the prevention of hemodialysis tunneled cuffed catheter thrombosis. J Vascular Access 7: 118–122, 2006 [DOI] [PubMed] [Google Scholar]
- 29.Abdul-Rahman IS, Al-Howaish AK: Warfarin versus aspirin in prevention tunneled hemodialysis catheter thrombosis: A prospective randomized study. Hong King J Nephrol 9: 23–30, 2007 [Google Scholar]
- 30.Ethier J, Bragg-Gresham JL, Piera L, Akizawa T, Asano Y, Mason N, Gillespie BW, Young EW: Aspirin prescription and outcomes in hemodialysis patients: The dialysis outcomes and practice patterns study (DOPPS). Am J Kidney Dis 50: 602–611, 2007 [DOI] [PubMed] [Google Scholar]
- 31.Holden RM, Harman GJ, Wang M, Holland D, Day AG: Major bleeding in hemodialysis patients. Clin J Am Soc Nephrol 3: 105–110, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.To AC, Yehia M, Collins JF: Atrial fibrillation in hemodialysis patients: Do the guidelines apply? Nephrology 12: 441–447, 2007 [DOI] [PubMed] [Google Scholar]
- 33.Fox KA, Mehta SR, Peters R, Zhao F, Lakkis N, Gersh BJ, Yusuf S; Clopidogrel in Unstable angina to prevent Recurrent ischemic Events Trial: Benefits and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: The Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) Trial. Circulation 7; 110: 1202–1208, 2004 [DOI] [PubMed] [Google Scholar]
- 34.Becker RC, Meade TW, Berger PB, Ezekowitz M, O'Connor CM, Vorchheimer DA, Guyatt GH, Mark DB, Harrington RA; American College of Chest Physicians: The primary and secondary prevention of coronary artery disease: American college of chest physicians evidence-based clinical practice guidelines. Chest 133: 776–814, 2008 [DOI] [PubMed] [Google Scholar]
- 35.Casterella PJ, Tcheng JE: Review of the 2005 American college of cardiology, American heart association, and society for cardiovascular interventions guidelines for adjunctive pharmacologic therapy during percutaneous coronary interventions: Practical implications, new clinical data, and recommended guideline revisions. Am Heart J 155: 781–790, 2008 [DOI] [PubMed] [Google Scholar]
- 36.Saran R, Dykstra DM, Wolfe RA, Gillespie B, Held PJ, Young EW; Dialysis Outcomes and Practice Patterns Study: Association between vascular access failure and the use of specific drugs: The Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis 40: 1255–1263, 2002 [DOI] [PubMed] [Google Scholar]
- 37.Best PJ, Steinhubl SR, Berger PB, Dasgupta A, Brennan DM, Szczech LA, Califf RM, Topol EJ; CREDO Investigators: The efficacy and safety of short- and long-term dual antiplatelet therapy in patients with mild or moderate chronic kidney disease: Results from the Clopidogrel for the Reduction of Events during Observation (CREDO) Trial. Am Heart J 155: 687–693, 2008 [DOI] [PubMed] [Google Scholar]
- 38.Keltai M, Tonelli M, Mann JF, Sitkei E, Lewis BS, Hawken S, Mehta SR, Yusuf S; CURE Trial Investigators: Renal function and outcomes in acute coronary syndrome: Impact of clopidogrel. Eur J Cardiovasc Prev Rehabil 14: 312–318, 2007 [DOI] [PubMed] [Google Scholar]
- 39.Lohr J, Schwab S: Minimizing hemorrhagic complications in dialysis patients. J Am Soc Nephrol 2: 961–975, 1991 [DOI] [PubMed] [Google Scholar]
- 40.Biggers JA, Remmers AR, Jr, Glassford DM, Sarles HE, Lindley JD, Fish JC: The risk of anticoagulation in hemodialysis patients. Nephron 18: 109–113, 1977 [DOI] [PubMed] [Google Scholar]
- 41.Vázquez E, Sánchez-Perales C, García-Cortes MJ, Borrego F, Lozano C, Guzmán M, Gil JM, Liébana A, Pérez P, Borrego MJ, Pérez V: Ought dialysis patients with atrial fibrillation be treated with oral anticoagulants? Int J Cardiol 87: 135–139, 2003 [DOI] [PubMed] [Google Scholar]
- 42.Foley RN, Curtis BM, Parfrey PS: Hemoglobin targets and blood transfusions in hemodialysis patients without symptomatic cardiac disease receiving erythropoietin therapy. Clin J Am Soc Nephrol 3: 1669–1675, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wells G, Shea B, O'Connell D, Peterson J, Welch V, Losos M, Tugwell P: The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. Proceedings of the 3rd Symposium on Systematic Reviews. Beyond the basics: Improving quality and impact; 2000 July 3–5, Oxford. Oxford, 2000 [Google Scholar]
- 44.Steinhubl SR, Bhatt DL, Brennan DM, Montalescot G, Hankey GJ, Eikelboom JW, Berger PB, Topol EJ; CHARISMA Investigators: On behalf of the CHARISMA investigators: Aspirin to prevent cardiovascular disease: The Association of Aspirin Dose and Clopidogrel With Thrombosis and Bleeding. Ann Intern Med 150: 379–386, 2009 [DOI] [PubMed] [Google Scholar]
- 45.Chan KE, Lazarus JM, Thadhani R, Hakim RM: Anticoagulant and antiplatelet usage associates with mortality among hemodialysis patients. J Am Soc Nephrol 20: 872–881, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]