Ticagrelor versus Clopidogrel in Patients with Acute Coronary Syndromes
2009 · paper · Clinical trials
A double-blind trial of 18,624 patients hospitalized with acute coronary syndrome, comparing ticagrelor with clopidogrel. Ticagrelor reduced the composite of vascular death, heart attack or stroke (9.8% versus 11.7%) without raising overall major bleeding, though non-procedure-related bleeding rose.
4 findings
Substance Supplementary Table S2, 'Anxiety disorder' row, Ticagrelor No COPD column, prints "56/1153 (0.9)"; 56/1153 is 4.9%, so the percentage should read 4.9.
- p. 8 Table S2, row 'Anxiety disorder', Ticagrelor No COPD column
56/1153 (0.9)
Judge's note: I checked the crop and the full images of supp-01 pages 7 and 8. Every cell in the Ticagrelor No COPD column uses the denominator 1153, the column's total number of dyspnea episodes, and every other percentage in that column matches its count/1153. The column's counts add up to exactly 1153 when anxiety disorder is 56, so the count is right and the printed percentage (0.9) is wrong. No other denominator in the table gives 0.9% for 56. The fix is evident: 4.9, which is probably a single-digit slip. No erratum is among the files.
Second check: another model agreed
Report a problem with this finding
Substance Table 4, clopidogrel column: 'Fatal bleeding 23/9186 (0.3)', but 'Nonintracranial fatal bleeding 21/9186' plus intracranial 'Fatal 1/9186' gives 22; the ticagrelor column sums exactly (9 + 11 = 20).
- p. 10 Table 4, Fatal bleeding, Clopidogrel column
23/9186 (0.3)
- p. 10 Table 4, Nonintracranial fatal bleeding, Clopidogrel
21/9186 (0.3)
- p. 10 Table 4, Intracranial bleeding, Fatal, Clopidogrel
1/9186 (0.01)
Only the authors can say what is right here.
Judge's note: Verified in the Table 4 image on page 10. The Results text on page 9 repeats 9 vs. 21 nonintracranial and 11 vs. 1 intracranial fatal bleeds. No footnote or text mentions fatal bleeds of unknown site or another counting rule. A patient with more than one event would make the parts add to more than the total, not less. Only the authors can say which count is correct.
Second check: another model agreed
Report a problem with this finding
Reference or label References 17 and 20 are the same entry ('Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy … the PCI-CURE study. Lancet 2001;358:527-33'); ref 20 is cited in the '20-25' group on higher loading dose and invasive/noninvasive management.
- p. 13 Reference 17
Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy
- p. 13 Reference 20
20. Mehta SR, Yusuf S, Peters RJ, et al.
- p. 13 References 17 and 20
20. Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients undergoing percutaneous coronary intervention: the PCI-CURE study. Lancet 2001;358:527-33.
- p. 10 Discussion, top left
noninvasive management was planned.20-25
Only the authors can say what is right here.
Judge's note: In the page 13 image, entries 17 and 20 are word-for-word identical, page range included. A numbered list should not hold the same source twice, so the source meant for ref 20 may be missing; only the authors can say which it is.
Second check: another model agreed
Report a problem with this finding
Reference or label Discussion: mortality reductions 'with clopidogrel in patients who had myocardial infarction with ST-segment elevation.22' cites ref 22 (Bonello et al., 'Tailored clopidogrel loading dose according to platelet reactivity monitoring to prevent acute and subacute stent thrombosis').
- p. 11 Discussion, paragraph 2
clopidogrel in patients who had myocardial infarction with ST-segment elevation.22
- p. 13 Reference 22
Tailored clopidogrel loading dose according to platelet reactivity monitoring to prevent acute and subacute stent thrombosis
- p. 13 References, entry 22
Bonello L, Camoin-Jau L, Armero S, et al. Tailored clopidogrel loading dose according to platelet reactivity monitoring to prevent acute and subacute stent thrombosis.
- p. 13 Reference 22
Tailored clopidogrel loading dose ac-cording to platelet reactivity monitoring
Only the authors can say what is right here.
Judge's note: Checked the page 11 sentence and ref 22 on page 13. By its printed title, ref 22 is a loading-dose and stent-thrombosis study, not a STEMI mortality trial. Ref 19 (Chen ZM et al., clopidogrel added to aspirin in 45,852 patients with acute MI) looks like the intended source, but only the authors can confirm which reference was meant.
Second check: another model agreed
Report a problem with this finding
9 minor (typos and the like; most checked by the judge only)
Wording Methods list the secondary end point 'death from cardiovascular causes alone', but the end point defined (End Points: 'Death from vascular causes was defined as death from cardiovascular causes or cerebrovascular causes…'), tested and reported (abstract, Results, Table 3) is 'Death from vascular causes'.
- p. 3 Statistical Analysis, list of secondary end points
death from cardiovascular causes alone
- p. 3 End Points
Death from vascular causes was defined as death from cardiovascular causes or cerebrovascular causes
- p. 8 Table 3
Death from vascular causes
Only the authors can say what is right here.
Judge's note: Checked page 3 (both columns) and Table 3 on page 8. The paper defines vascular death as broader than cardiovascular death, and the hierarchical-testing list names 'cardiovascular causes alone'. Table 3 and the abstract report 'Death from vascular causes' (353 vs. 442, P=0.001). This is probably a terminology slip, but because the paper defines the two terms differently, the authors should say which end point was prespecified.
Second check: another model agreed
Report a problem with this finding
Wording Methods: consistency over time 'was assessed by determining the relative risk ratios for the periods from randomization to 30 days and from 31 to 360 days', whereas Table 3 reports hazard ratios (Cox regression) for these periods.
- p. 5
relative risk ratios for the periods from randomization
Judge's note: Page 5 image shows 'relative risk ratios'. Table 3 (page 8) labels the column 'Hazard Ratio' and footnote † says Cox regression was used. This is loose wording rather than a different analysis.
Second check: another model agreed
Report a problem with this finding
Wording Supplementary Table S2 gives the ticagrelor group as '(N=9233)' (550 COPD + 8683 no COPD), while main Table 4 and Fig. 2 use 9235 (e.g., '1270/9235 (13.8)'); the clopidogrel N (9186) matches.
- p. 7 Table S2 header, Ticagrelor
(N=9233)
- p. 10 Table 4, Ticagrelor denominators
1270/9235 (13.8)
Only the authors can say what is right here.
Judge's note: Checked the supp page 7 image and Table 4 on page 10. The event counts match (117+1153=1270; 71+650=721). The supplement gives no note on missing COPD status or other exclusions. The 2-patient difference may be legitimate, but it is unexplained.
Second check: another model agreed
Report a problem with this finding
Wording Discussion: 'The treatment effects were the same in the short term (days 0 to 30) and in the longer term (days 31 to 360)', while Table 3 gives 0.88 (0.77–1.00) for 'days 1–30' and 0.80 (0.70–0.91) for 'days 31–360', and no test of interaction with time is reported.
- p. 10 Discussion, top left
treatment effects were the same in the short term (days 0 to 30) and in the longer term (days 31 to 360)
- p. 8 Table 3, event rate rows
0.88 (0.77–1.00)
- p. 8 Table 3, days 31–360
0.80 (0.70–0.91)
Only the authors can say what is right here.
Judge's note: Checked the page 10 Discussion and Table 3 on page 8. The point estimates differ, the table labels the period 'days 1–30', not 'days 0 to 30', and no time-interaction test is shown. 'The same' says more than the table shows; 'consistent' would match the data.
Second check: another model agreed
Report a problem with this finding
Typo Supplementary Table S2, Heart failure (non-cardiac etiology), clopidogrel COPD: '0/71 (0.0%)' carries a % sign that no other cell has.
- p. 7 Table S2, Heart failure (non-cardiac etiology), Clopidogrel COPD
0/71 (0.0%)
Judge's note: The supp page 7 image shows '(0.0%)'. All other cells print the percentage without a % sign, as the header '(%)' already specifies.
Second check: none (judge only)
Report a problem with this finding
Typo Table 4, Non–CABG-related major bleeding, TIMI criteria: '1.25 (1.03, 1.53)' uses a comma where every other interval uses an en dash.
- p. 10 Table 4, Non–CABG-related major bleeding, TIMI criteria
1.25 (1.03, 1.53)
Judge's note: The page 10 image shows the comma. Every other interval in Tables 3 and 4 uses the form x–y.
Second check: none (judge only)
Report a problem with this finding
Typo Disclosures (Dr. Mahaffey): 'Daichii Sankyo', spelt 'Daiichi Sankyo' everywhere else in the same disclosure paragraph.
- p. 12 Disclosures, Dr. Mahaffey
Daichii Sankyo
Judge's note: The page 12 image shows 'Daichii Sankyo' in Dr. Mahaffey's entry and 'Daiichi Sankyo' in the Scirica, Steg and Storey entries.
Second check: none (judge only)
Report a problem with this finding
Typo Disclosures (Dr. Harrington): 'grant support from Millenium Pharmaceuticals'.
- p. 12 Disclosures, Dr. Harrington
Millenium Pharmaceuticals
- p. 12 Disclosures, Dr. Harrington
Millenium
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the company's name is spelt 'Millennium Pharmaceuticals'. The page 12 image shows 'Millenium'.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 30 title: 'Effects of fondaparinux, on mortality and reinfarction…' has a stray comma.
- p. 13 Reference 30
Effects of fondaparinux, on mortality
Judge's note: The page 13 image shows the comma after 'fondaparinux', which breaks the grammar of the title.
Second check: none (judge only)