Strategies for Multivessel Revascularization in Patients with Diabetes
2012 · paper · Clinical trials
A randomized trial of 1900 patients with diabetes and multivessel coronary disease, comparing drug-eluting stents with bypass surgery. Five-year death, heart attack or stroke occurred in 26.6% after stenting and 18.7% after surgery; stroke was more frequent with surgery.
2 findings
Substance Results: "All the procedural myocardial infarctions in the trial were non–Q-wave events", but the Methods define MI in the first 30 days after any revascularization as "the presence of new Q waves in 2 or more contiguous leads on electrocardiography".
- p. 6 Results, Primary Outcome, first lines of left column below Figure 1
All the procedural myocardial infarctions in the trial were non–Q-wave events.
- p. 3 Trial Outcomes, MI definition
myocardial infarction was defined as the presence of new Q waves in 2 or more contiguous leads
- p. 3 Methods, Trial Outcomes, MI definition
myocardial infarction was defined as the presence of new Q waves in 2 or more contiguous leads on electrocardiography
- p. 7 Discussion, right column
we used even more stringent criteria for the definition of periprocedural myocardial infarction.
Only the authors can say what is right here.
Judge's note: Checked p3 (MI definition) and p6 (Results). The Methods give only a Q-wave definition for MI within 30 days of a procedure; the enzyme-based criteria apply only after 30 days. The p7 Discussion adds that 'even more stringent criteria' were used for periprocedural MI. Under that definition a procedural MI must be a Q-wave event, so 'non–Q-wave' contradicts the Methods. 'Q-wave' was probably intended, but only the authors can confirm.
Second check: another model agreed
Report a problem with this finding
Substance Table 2 gives 2-year cardiovascular death as "9 (0.9)" (PCI) and "12 (1.3)" (CABG) against 5-year "73 (10.9)" and "52 (6.8)"; with all-cause death at "62 (6.7)"/"57 (6.3)" (2 yr) and "114 (16.3)"/"83 (10.9)" (5 yr), this implies more CV deaths than all deaths between years 2 and 5.
- p. 5 Table 2, Cardiovascular death row, 2 years
9 (0.9)
- p. 5 Table 2, Cardiovascular death row, 2 years, CABG
12 (1.3)
- p. 5 Table 2, Death from any cause row
62 (6.7)
- p. 6 Secondary Outcomes
Rates of cardiovascular death (63.7% of all deaths)
Only the authors can say what is right here.
Judge's note: Checked Table 2 on the p5 image. Cardiovascular deaths are a subset of all deaths, yet between years 2 and 5 the table adds 64 CV deaths against 52 all-cause deaths in the PCI group, and 40 against 26 in the CABG group. The 2-year CV figures also clash with Table 3 (32 and 38 deaths from any cause by 12 months) and with the text on p6 that CV deaths were 63.7% of all deaths. The 2-year CV death entries look wrong as printed. The correct values are not in the files.
Second check: another model agreed
Report a problem with this finding
5 minor (typos and the like; most checked by the judge only)
Wording Methods describe "the Rankin scale (ranging from 1 to 5, with higher scores indicating greater disability)".
- p. 3 Trial Outcomes
(ranging from 1 to 5, with higher scores indicat-
- p. 3 Trial Outcomes
ranging from 1 to 5, with higher scores indicat-
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the modified Rankin scale runs from 0 to 6 (or 0 to 5 among survivors). Page 3 does print 'ranging from 1 to 5'. Inside the paper nothing contradicts this: the Results cut-off 'more than 1' fits a 1-to-5 range too. So this is a query to the authors about the stated range.
Second check: another model agreed
Report a problem with this finding
Wording Table 1 gives "No. of lesions 5.65±2.16 / 5.74±2.19", but the ¶ denominators ("323/5564", "329/5662"; "1242/5561", "1177/5640"), which are said to be the total number of lesions in 949 PCI and 939 CABG patients, give about 5.86 and 6.03 lesions per patient.
- p. 4 Table 1, No. of lesions row
5.65±2.16
- p. 4 Table 1, Chronic total occlusion row
323/5564 (5.8)
- p. 4 Table 1, footnote ¶
based on the total number of lesions in 949 patients in the PCI group and 939 patients in the CABG group
Only the authors can say what is right here.
Judge's note: Checked Table 1 on the p4 image. 5.65×949 ≈ 5362 lesions, fewer than the 5564 the footnote gives as total lesions. Even dividing by the full 953 or 947 patients still gives about 5.84 or 5.98. The 'No. of lesions' mean may count lesions under a different definition (for example significant stenoses only) from the lesion denominators, but the table does not say so. Only the authors can say which definition each row uses.
Second check: another model agreed
Report a problem with this finding
Wording "The excess of strokes in the CABG group occurred in the first 30 days after randomization" and "The higher relative risk of stroke among patients undergoing CABG was evident only early", while Table 2 shows strokes rising from 14 vs 24 at 2 years to 20 vs 37 at 5 years.
- p. 6 Results, stroke paragraph
The excess of strokes in the CABG group occurred in the first 30 days after randomization.
- p. 9 Discussion, second paragraph
was evident only early in the postprocedural period
- p. 5 Table 2, Stroke row
14 (1.5) 24 (2.7) 20 (2.4) 37 (5.2)
Only the authors can say what is right here.
Judge's note: Checked Table 2 (p5), Table 3 (p7), the p6 text and the p9 Discussion. The CABG excess is 13 at 30 days (Table 3), 10 at 2 years and 17 at 5 years. Between 2 and 5 years CABG added 13 strokes against 6 for PCI. 'Only early' therefore overstates the tabulated counts, though the numbers are small. The 30-day counts the sentence relies on are also timed from the procedure, not from randomization. A wording query.
Second check: another model agreed
Report a problem with this finding
Typo Results say "An NIH Stroke Scale score of more than 4 (severely disabling)" while the Methods define ">4 indicating a disabling stroke".
- p. 6
An NIH Stroke Scale score of more than 4 (severely disabling)
Judge's note: Checked p3 and p6. The threshold is the same in both places; only the label changes ('disabling' vs 'severely disabling'). A minor consistency point in wording.
Second check: none (judge only)
Report a problem with this finding
Typo Disclosures: "serving as an expert witness in medical malpractices cases regarding medical management" should read "medical malpractice cases".
- p. 9 Disclosures, Dr. Weinberger
expert witness in medical malpractices cases re-
Judge's note: Checked the p9 image: 'malpractices cases' is printed there. Dr. Desai's entry on the same page uses 'a malpractice case'.
Second check: none (judge only)