Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy
2019 · paper · Clinical trials
A randomized trial of canagliflozin versus placebo in 4401 people with type 2 diabetes and kidney disease. The canagliflozin group had a 30% lower risk of kidney failure, doubled creatinine, or kidney or cardiovascular death over a median 2.62 years.
0 findings
None that passed a second check.
5 minor (typos and the like; most checked by the judge only)
Wording Table 2 footnote: "P values are reported only for outcomes that were included in the hierarchical-testing strategy", yet the non-hierarchical components doubling of serum creatinine (<0.001) and end-stage kidney disease (0.002) carry P values.
- p. 7 Table 2 footnote *
P values are reported only for outcomes that were included in the hierarchical-testing strategy
- p. 6 Table 2, Doubling of serum creatinine level and End-stage kidney disease rows
0.60 (0.48–0.76)
- p. 3 Outcomes, list of hierarchical secondary outcomes
Secondary outcomes that were planned for sequential hierarchical testing were specified in the following order
Only the authors can say what is right here.
Judge's note: The Methods on p3 list the primary outcome and seven ordered secondary outcomes and state 'All other efficacy outcomes were exploratory.' Doubling of serum creatinine and end-stage kidney disease, as separate outcomes, are not in that list. Table 2 (p6 image) nevertheless prints P values for them, and the abstract and Results repeat P=0.002 and P<0.001. The paper gives no stated exception, so the footnote's rule does not match the table. The authors should clarify the status of these P values.
Second check: another model agreed
Report a problem with this finding
Wording Text says "Rates of adverse events and serious adverse events were similar overall", but Table 2 gives HRs of 0.87 (0.82–0.93) and 0.87 (0.79–0.97), both intervals excluding 1.
- p. 7 Effects on Safety Outcomes, first sentence
Rates of adverse events and serious adverse events were similar overall in the canagliflozin group and the placebo group
- p. 6 Table 2, Safety rows
0.87 (0.82–0.93)
Judge's note: Checked p7 text and the Table 2 safety rows on p6. Both intervals lie entirely below 1, so the word 'similar' understates what the table shows. This is wording only (S3). The second reviewer's claim that HR 0.87 is misprinted is not supported. The interval is symmetric about 0.87 on the log scale (sqrt(0.82*0.93) ≈ 0.873). The crude rates per 1000 patient-years appear to use total follow-up, not time to first event: 1784/351.4*1000 ≈ 5077 patient-years, about total follow-up for 2200 patients at a median of 2.62 years, even though about 81% had an event. A Cox time-to-first-event HR is therefore a different quantity from the crude rate ratio and need not match it.
Second check: another model agreed
Report a problem with this finding
Wording Figure 3 caption: "Panel B shows the change from the screening level in the estimated GFR", whereas the Panel B title reads "Change from Baseline in Estimated GFR" and shows Baseline values 56.4 and 56.0.
- p. 10 Figure 3 caption
Panel B shows the change from the screening level in the estimated GFR
- p. 10 Figure 3 Panel B title
Change from Baseline in Estimated GFR
- p. 10 Figure 3 legend
change from the screening level in the estimated GFR
- p. 10 Figure 3 caption, Panel B
Panel B shows the change from the screening level in the estimated GFR in the on-treat-ment population.
Only the authors can say what is right here.
Judge's note: Checked p10. The caption and the panel title give different reference points. Elsewhere the paper keeps the terms apart: stratification uses eGFR category at screening, and the Methods define baseline as the average of the randomization and prerandomization values. Which reference the plotted changes use is therefore ambiguous. Wording issue only.
Second check: another model agreed
Report a problem with this finding
Typo Table 2: the section-heading row "Safety‡" carries a stray "NA" in the P Value column, unlike the other heading rows (Efficacy, Secondary outcomes, Cancer).
- p. 6 Table 2, 'Safety‡' heading row
Safety‡
Judge's note: In the upright image of p6, the 'Safety‡' row has no data in any column except 'NA' under P Value. The heading rows Efficacy, Secondary outcomes and Cancer have nothing in that column. The entry is redundant and cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Affiliations: "Department of Medicine (B.M.B), Brigham and Women’s Hospital" lacks the final period; elsewhere the initials appear as "B.M.B.".
- p. 12 Appendix, affiliations
Department of Medicine (B.M.B), Brigham and Women’s Hospital
Judge's note: In the p12 crop the initials appear as 'B.M.B.' twice in the same paragraph and as '(B.M.B)' once. Cosmetic punctuation slip.
Second check: none (judge only)