Dapagliflozin in Patients with Chronic Kidney Disease
2020 · paper · Clinical trials
A randomized trial of dapagliflozin versus placebo in 4304 people with chronic kidney disease, with or without diabetes. The primary composite of kidney decline, kidney failure, or kidney or cardiovascular death occurred in 9.2% versus 14.5%, and effects were similar regardless of diabetes.
2 findings
Reference or label Figure 1 legend: "Included in these analyses are all the participants who had undergone randomization and received at least one dose of dapagliflozin or placebo", but every panel starts with 2152 at risk per group (all randomized), and the Methods say the efficacy analysis used "the intention-to-treat population, which included all the participants who had undergone randomization".
- p. 7 Figure 1 legend
Included in these analyses are all the participants who had undergone ran-
- p. 7 Figure 1, No. at Risk rows, month 0, all panels
Dapagliflozin 2152
- p. 3 Statistical Analysis, last paragraph
The primary efficacy analysis was based on
- p. 6 Table 2, Safety outcomes denominators
118/2149 (5.5)
Judge's note: Checked the page 7 image (legend and No. at Risk rows: 2152 in both groups in all four panels), page 3/4 (intention-to-treat definition; the dosed set is defined there for safety data only) and Table 2 on page 6 (efficacy denominators 2152, safety denominators 2149, footnote ‡ gives the dosed definition for safety analyses). The legend describes the safety set (2149 per group), not the population plotted. The population label is wrong; the data are unaffected.
Second check: another model agreed
Report a problem with this finding
Wording "The incidences of adverse events and serious adverse events were similar overall in the dapagliflozin and placebo groups (Table 2)", but Table 2 gives serious adverse events as "633/2149 (29.5)" versus "729/2149 (33.9)", P = 0.002, and has no row for adverse events overall.
- p. 9 Safety Outcomes and Adverse Events, first sentence
The incidences of adverse events and serious ad-
- p. 6 Table 2, Any serious adverse event
633/2149 (29.5)
- p. 6 Table 2, Any serious adverse event, P Value
0.002
- p. 9 Results, Safety Outcomes and Adverse Events, first sentence
verse events were similar overall in the dapa-
Only the authors can say what is right here.
Judge's note: Checked the page 9 image and the upright Table 2 image. Percentages recompute correctly; the difference is 4.4 percentage points with a printed P of 0.002, which the text calls similar. Table 2 lists only discontinuations due to adverse events, serious adverse events and selected events of interest, and the Methods say only selected adverse-event data were collected, so the cited table does not show an overall adverse-event incidence. Wording overstates similarity; no explanation found elsewhere in the article.
Second check: another model agreed
Report a problem with this finding
6 minor (typos and the like; most checked by the judge only)
Wording Discussion: "hypoglycemic episodes did not occur in participants without diabetes"; the Results say only that "Neither diabetic ketoacidosis nor severe hypoglycemia was observed in participants without type 2 diabetes", and only "major hypoglycemia" was collected.
- p. 10 Discussion, safety paragraph
hypoglycemic episodes did not occur in partici-
- p. 9 Safety Outcomes and Adverse Events
dosis nor severe hypoglycemia was observed in
- p. 3 Outcomes, adverse events of interest
events, major hypoglycemia, bone fractures, am-
- p. 9 Results, Safety Outcomes and Adverse Events
Neither diabetic ketoaci-
Judge's note: Checked pages 3, 9, 10 and Table 2. The Methods list major hypoglycemia among the selected adverse events collected; Table 2 reports only major hypoglycemia (footnote defines it); the Results say severe hypoglycemia. The Discussion drops the qualifier, which is broader than the data collected. Also the same event is called major and severe. Probably shorthand; minor wording point.
Second check: another model agreed
Report a problem with this finding
Wording Figure 1 legend: "The graphs are truncated at 32 months (the point at which <15% of participants remained at risk)", but at 32 months panels C and D show 360/384 and 379/398 of 2152 at risk (16.7% to 18.5%).
- p. 7 Figure 1 legend
The graphs are truncated at 32 months (the point at which <15%
- p. 7 Figure 1C, No. at Risk at 32 months
360
- p. 7 Figure 1D, No. at Risk at 32 months
398
Judge's note: Read the No. at Risk rows on the page 7 image. Panels A and B: 270 and 309 (12.5% and 14.4%), under 15%. Panels C and D: 360, 384, 379, 398, all above 15%. The statement holds for the primary outcome and panel B only; the legend applies it to all graphs. Probably the rule was set on the primary outcome; minor.
Second check: another model agreed
Report a problem with this finding
Typo Coauthor's surname is printed "Stefánsson" in the byline and reference 20 ("Pollock C, Stefánsson B, Reyner D") but "Stefansson" in references 14 ("Heerspink HJL, Stefansson BV") and 15 ("Wheeler DC, Stefansson BV").
- p. 1 Author byline
Bergur V. Stefánsson, M.D.
- p. 11 Reference 14
Heerspink HJL, Stefansson BV, Cher-
- p. 11 Reference 15
Wheeler DC, Stefansson BV, Batiushin
- p. 11 Reference 20
Pollock C, Stefánsson B, Reyner D, et
Judge's note: Checked the byline crop and the page 11 image: the acute accent is present in the byline and reference 20 and absent in references 14 and 15. Cosmetic; references 14 and 15 may reproduce the spelling used by the cited journal, so either form could be legitimate.
Second check: none (judge only)
Report a problem with this finding
Typo Appendix affiliation reads "the Department Clinical Pharmacy and Pharmacology, University of Groningen"; the word "of" is missing.
- p. 10 Appendix, authors' affiliations, first line
the Department Clinical Pharmacy and Pharmacology
- p. 1 Correspondence note, left margin
Dr. Heerspink at the Department of Clini-
Judge's note: Confirmed on the page 10 image: "the Department Clinical Pharmacy and Pharmacology". Other departments in the same list use "Department of", and the reprint address on page 1 (text layer) reads "Department of Clinical Pharmacy and Pharmacology".
Second check: none (judge only)
Report a problem with this finding
Typo "Diabetic ketoacidosis was not reported in any participants who received dapagliflozin and in two participants who received placebo."
- p. 9 Safety Outcomes and Adverse Events, second sentence
who received dapagliflozin and in two participants
- p. 6 Table 2, Any definite or probable diabetic ketoacidosis, placebo
2/2149 (<0.1)
Judge's note: Page 9 image checked; sentence is printed as quoted. The negation grammatically covers both halves. Table 2 shows the intended meaning (0/2149 with dapagliflozin, 2/2149 with placebo); a verb is missing (e.g. "and was reported in two participants"). Wording slip only.
Second check: none (judge only)
Report a problem with this finding
Typo "Our trial has some limitations. First, the trial was stopped on the basis of a recommendation from the independent data monitoring committee." No "Second" or further ordinal follows.
- p. 10 Discussion, limitations paragraph
Our trial has some limitations. First, the trial
Judge's note: Read the whole paragraph on the page 10 image. After "First" the paragraph moves to the initial eGFR dip and the lack of post-trial eGFR values without marking them as a further limitation. Cosmetic; the unmarked eGFR passage may be the intended second limitation.
Second check: none (judge only)