Randomized Assessment of Rapid Endovascular Treatment of Ischemic Stroke
2015 · paper · Clinical trials
A randomized trial of rapid endovascular treatment plus standard care versus standard care alone in 316 stroke patients with a proximal occlusion, small infarct core and good collaterals. Functional independence was 53.0% versus 29.3%, and mortality was lower; the trial stopped early for efficacy.
4 findings
Substance Table 3 footnote: "Differences (intervention group − control group) are shown as percentage points", but the Death row (10.4 vs 19.0) prints "8.6 (0.8 to 16.6)" and the large or malignant MCA stroke row (4.8 vs 10.7) prints "5.8 (0.1 to 11.7)", both positive; by the footnote's definition they are −8.6 and −5.8. The symptomatic ICH row, "1.0 (−2.9 to 4.8)", does follow the footnote.
- p. 8 Table 3, row Death, Difference column
8.6 (0.8 to 16.6)
- p. 8 Table 3, row Large or malignant middle-cerebral-artery stroke, Difference column
5.8 (0.1 to 11.7)
- p. 8 Table 3, footnote *
Differences (intervention group − control group) are shown as percentage points.
- p. 8 Table 3, row Symptomatic intracerebral hemorrhage, Difference column
1.0 (−2.9 to 4.8)
Judge's note: Read Table 3 on the crop and page 8 image. No minus sign is printed on 8.6, 0.8, 16.6, 5.8, 0.1 or 11.7, while the minus on −2.9 is clearly printed, so this is not an extraction loss. Intervention minus control is 10.4−19.0 = −8.7 (from counts −8.68) and 4.8−10.7 = −5.8 (from counts −5.82); the third row, 3.6−2.7 = +1.0, is in the footnote's direction. The column therefore mixes two directions against its own footnote. The adjacent percentages and rate ratios below 1 let a careful reader recover the direction, but as printed the sign of two differences is opposite to the stated definition. Separately and minor: the death difference from the counts is 8.68, which rounds to 8.7 (interval midpoint 8.7), while 8.6 is printed.
Second check: another model agreed · checked by hand
Report a problem with this finding
Reference or label Results: "(For details on the TICI scale, see Table 3 in the Supplementary Appendix.)" — elsewhere the same content is cited as "Table S3 in the Supplementary Appendix" (Table 2 footnotes ‖ and **, and the modified AOL sentence on the same page); "Table 3" is the article's own serious-adverse-events table.
- p. 8 Results, end of left column
Table 3 in the Supplementary Appendix.)
- p. 7 Table 2 footnote ‖
indicates successful reperfusion (see Table S3 in the Supplementary Appendix)
- p. 8 Results, reperfusion paragraph
see Table 3 in the Supplementary Appendix.)
- p. 7 Table 2 footnote ‖
see Table S3 in the Supplementary Appendix
Judge's note: Read the sentence on the page 8 image and the Table 2 footnotes on the page 7 image. Table 2 footnote ‖ cites Table S3 for the TICI score, footnote ** and the next paragraph cite Table S3 for the AOL scale, and every other supplement pointer in the article carries the S prefix. The supplement is not among the files, so its numbering was not checked directly, but the article is internally inconsistent and the evident fix is 'Table S3'.
Second check: another model agreed
Report a problem with this finding
Wording Table 1, location of occlusion, intervention group: "45/163 (27.6)", "111/163 (68.1)", "6/163 (3.7)" sum to 162 of a denominator of 163, and the footnote "In one participant in the intervention group, the location of the occlusion on CTA was not determined by the core laboratory" would give 164 of N=165, not 163.
- p. 4 Table 1, Location of occlusion
45/163 (27.6)
- p. 5 Table 1 footnote ‖
In one participant in the intervention group, the location of the occlusion on CTA was not determined
- p. 5 Table 1 footnote ‖
In one participant in the intervention group, the location of the occlusion on CTA was not determined by the core lab-
Only the authors can say what is right here.
Judge's note: Read Table 1 on the page 4 image and the footnote on the page 5 image. Intervention counts 45+111+6 = 162 against a printed denominator of 163; control counts 39+105+3 = 147 match their denominator. The footnote accounts for one undetermined participant in the intervention group (165-1 = 164) and says nothing of the three missing in the control group (150-147). The percentages are computed on 163. No 'other location' row or further explanation was found in the article (the text mentions 14 participants with inappropriate target-vessel occlusion, but does not tie them to this table). Either a category or a count is missing, or the denominator/footnote is off; only the authors can say which.
Second check: another model agreed
Report a problem with this finding
Wording Results: "There was no evidence of heterogeneity of effect across any of the prespecified subgroups ... or according to the presence or absence of cervical carotid occlusion", while the Figure 2 legend prints "For cervical carotid occlusion, P = 0.049 for interaction by the Wald test" and the Methods state "P values of less than 0.05 were considered to indicate statistical significance".
- p. 7 Results, Secondary outcomes and subgroup analyses
There was no evidence of heterogeneity of effect across any of the prespecified subgroups
- p. 7 Results, same sentence
or according to the presence or absence of cervical carotid occlusion
- p. 9 Figure 2 legend
For cervical carotid occlusion, P = 0.049 for interaction by the Wald test.
- p. 4 Statistical Analysis
P values of less than 0.05 were considered to indicate statistical significance
Only the authors can say what is right here.
Judge's note: Read the Results sentence on the page 7 image, Figure 2 and its legend on the page 9 image, and the Methods on the page 4 image. The legend prints P = 0.049 for the cervical carotid occlusion interaction (rows 9.6 (2.6–35.5) and 2.2 (1.4–3.3)), which is below the paper's stated 0.05 threshold with no multiplicity adjustment, yet the Results sentence names cervical carotid occlusion as showing no evidence of heterogeneity and does not mention this P value. This is a text-against-printed-value point, not a recomputation. There is also a labelling difference: the Results list of prespecified subgroups omits cervical carotid occlusion and time to randomization, while the Figure 2 legend calls all plotted subgroups prespecified. Query for the authors on the intended wording (or whether 0.049 is as intended).
Second check: another model agreed
Report a problem with this finding
6 minor (typos and the like; most checked by the judge only)
Wording Table 2, NIHSS score of 0–2 at 90 days: difference printed "28.4 (17.8–39.2)" while 79/153 (51.6) minus 31/134 (23.1) is 28.5, and the printed interval is centred on 28.5.
- p. 7 Table 2, NIHSS score of 0–2 row
28.4 (17.8–39.2)
Judge's note: Read Table 2 on the page 7 image. 79/153 = 51.634% and 31/134 = 23.134%; the difference is 28.4997, which rounds to 28.5, and the midpoint of 17.8–39.2 is 28.5. The other two difference rows agree with the counts (23.8 and 24.1). The discrepancy is 0.1 percentage point, sitting just under 28.5, so truncation or a floating-point rounding step would explain it; it changes nothing a reader would take away.
Second check: another model agreed
Report a problem with this finding
Typo Affiliations: "Neurosurgery (B.T.J.), University of Pittsburgh Medical Center" — no author has these initials; the byline gives "B. Jankowitz" and the full-name list "Brian Jankowitz, M.D." with no middle initial.
- p. 11 Appendix, affiliations
Neurosurgery (B.T.J.)
- p. 1 Byline
B. Jankowitz
- p. 10 Appendix, full names
Brian Jankowitz, M.D.
Only the authors can say what is right here.
Judge's note: Read the byline on the page 1 crop, the full-name list on the page 10 crop and the affiliations on the page 11 crop. 'B.T.J.' appears only in the affiliation key; the only matching author is Brian Jankowitz, listed twice without a middle initial. Either the key has a stray 'T' or the byline and name list omit a middle initial; the authors must say which.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 footnote ¶: "Scores ranges from 0 to 10, with higher scores indicating a smaller infarct core" — subject-verb disagreement; should read "Scores range".
- p. 5 Table 1 footnote ¶
Scores ranges from 0 to 10
- p. 5
Scores ranges from 0 to 10, with higher scores indicating a smaller infarct core
Judge's note: Seen on the page 5 image; the wording is in the printed footnote and is not an extraction artefact. The NIHSS footnote in the same table uses 'range'.
Second check: none (judge only)
Report a problem with this finding
Typo "100 of these 130 participants (77.0%) received a Solitaire stent" — 100/130 is 76.9%.
- p. 8 Results, left column
130 participants (77.0%) received a Solitaire stent
- p. 8 Results, treatments paragraph
100 of these 130 participants (77.0%) received a Solitaire stent
- p. 8 Results, Treatments
100 of these 130 participants (77.0%)
Judge's note: Read on the page 8 image. 100/130 = 76.92%, which is 76.9% to one decimal (77% as a whole number); 77.0 is not reachable by rounding. The neighbouring percentages (91.5, 72.7, 9.1, 86.1) are correctly rounded. A 0.1-point slip with no effect on meaning.
Second check: none (judge only)
Report a problem with this finding
Typo Discussion: "A total of 49 participants (15.5%) underwent randomization 6 or more hours after symptom onset" — 49/315 is 15.6%; 15.5% corresponds to a denominator of 316.
- p. 10 Discussion, left column
A total of 49 participants (15.5%)
Judge's note: Read on the page 10 crop. 49/315 = 15.56% (15.6%) and 49/316 = 15.51% (15.5%). Other whole-trial percentages use 315 (26 = 8.3%, 56 = 17.8%, 20 of 315 = 6.3%). The sentence is about participants who 'underwent randomization', and 316 underwent randomization, so a denominator of 316 could be intended; either form is defensible and the difference is 0.1 point.
Second check: none (judge only)
Report a problem with this finding
Typo Affiliations: "Colorado Neurological Institute, Engelwood (D.F.F.)" — the city name is printed "Engelwood"; the Colorado city is spelled Englewood.
- p. 11 Appendix, affiliations
Colorado Neurological Institute, Engelwood (D.F.F.)
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the Colorado city where the Colorado Neurological Institute is located is spelled 'Englewood'. The page 11 image prints 'Engelwood', so the quote is accurate and not an extraction artefact; the correct spelling cannot be verified from the files.
Second check: none (judge only)