A Randomized, Controlled Trial of Early versus Late Initiation of Dialysis
2010 · paper · Clinical trials
A randomized trial in Australia and New Zealand of 828 adults with advanced chronic kidney disease, comparing planned early versus late start of dialysis. Early initiation did not improve survival or clinical outcomes (37.6% vs 36.6% deaths).
1 finding
Substance Results: "During the recruitment phase of the trial (July 1, 2000, through November 14, 2006)" conflicts with "Between July 2000 and November 2008, a total of 828 patients ... were randomly assigned" in the same paragraph and in the abstract.
- p. 4 Results, Characteristics of the Patients
(July 1, 2000, through November 14, 2006)
- p. 4 Results, Characteristics of the Patients, first sentence
Between July 2000 and November 2008, a total of 828 patients
- p. 1 Abstract, Results
Between July 2000 and November 2008, a total of 828 adults
- p. 4 Results, Characteristics of the Patients
During the recruitment phase of the trial (July 1, 2000, through November 14, 2006)
Only the authors can say what is right here.
Judge's note: I checked p.4 and the abstract crop on p.1. Both say randomization ran from July 2000 to November 2008, and Figure 1 gives follow-up through November 14, 2009. The parenthetical explicitly calls the period ending in 2006 the trial's recruitment phase. Either the year is wrong or the 12,101-patient ANZDATA comparison covers a different window; only the authors can say which.
Second check: another model agreed
Report a problem with this finding
6 minor (typos and the like; most checked by the judge only)
Reference or label The Introduction cites refs 11-15 for "starting dialysis early may, in fact, be harmful.11-15", but the Discussion assigns ref 15 to "no effect on survival15" and only refs 11-14 to "potentially worsened survival".
- p. 2 Introduction
more recent observational data have suggested that starting dialysis early may, in fact, be harmful.11-15
- p. 7 Discussion, second paragraph
early initiation of dialysis had no effect on survival15 or potentially worsened survival.11-14
- p. 11 Reference 15
Early initiation of dialysis fails to prolong survival in patients with end-stage renal failure
Judge's note: I read p.2 (Introduction), p.7 (Discussion) and reference 15 on p.11 (Traynor et al., 'Early initiation of dialysis fails to prolong survival...'). The paper describes the same reference two different ways. The Introduction's range plausibly should be 11-14, although its looser grouping is defensible.
Second check: another model agreed
Report a problem with this finding
Wording The abstract and Methods say "32 centers in Australia and New Zealand", but the Appendix 'Study centers' list names 25 Australian and 8 New Zealand sites (33 entries; 34 hospitals if 'Blacktown/Westmead Hospitals' is counted as two).
- p. 1 Abstract, Background
conducted at 32 centers in Austra-
- p. 2 Methods, Patients
Patients were recruited at 32 centers in Australia
- p. 11 Appendix, Study centers
D.C. Harris, Blacktown/Westmead Hospitals, Sydney, NSW
- p. 1 Abstract, Background
conducted at 32 centers in Austra-lia and New Zealand
Only the authors can say what is right here.
Judge's note: I counted the sites on the p.11 page image. Australia: Alfred, Austin, Blacktown/Westmead, Cairns, Dubbo, John Hunter, Launceston, Liverpool, Monash, Nambour, Nepean, Orange, Princess Alexandra, Queen Elizabeth, Royal Adelaide, Royal Brisbane, Royal Hobart, Royal North Shore, Royal Perth, Sir Charles Gairdner, St George, St Vincents, Tamworth, Toowoomba, Wollongong (25). New Zealand: Auckland City, Dunedin, Middlemore, Palmerston North, Taranaki, Waikato, Wellington, Whangarei (8). The paper never says that a listed site enrolled nobody or that two sites were counted as one center. Only the authors can reconcile the count.
Second check: another model agreed
Report a problem with this finding
Wording The Discussion refers to "a post hoc analysis of survival with baseline GFR estimated according to the MDRD equation", yet the Methods describe only a post hoc comparison of MDRD GFR at the start of dialysis, and the Figure 3 legend calls its GFR MDRD subgroup one of "the prespecified subgroups".
- p. 9 Discussion, limitations paragraph
a post hoc analysis of survival with baseline GFR estimated according to the MDRD equation revealed no significant between-group differences in outcomes
- p. 4 Methods, Statistical Analysis
We also performed a post hoc analysis in which the GFR at the start of dialysis, as estimated with the use of the MDRD equation,25 was compared between
- p. 10 Figure 3 legend
according to each of the prespecified subgroups
Only the authors can say what is right here.
Judge's note: I checked p.4 (Statistical Analysis), p.9 (Discussion) and p.10 (Figure 3, which includes a GFR MDRD <9.5/>=9.5 subgroup, interaction P=0.58). The paper does not make clear whether the baseline-MDRD survival analysis was prespecified or post hoc.
Second check: another model agreed
Report a problem with this finding
Wording The Discussion says "The follow-up of patients was long and complete", while Figure 1 lists 43 patients who declined consent to the study extension, 14 who withdrew consent, 8 transferred to nonparticipating hospitals, 2 who emigrated and 2 lost to follow-up.
- p. 9 Discussion, limitations paragraph
The follow-up of patients was long and complete.
- p. 5 Figure 1
23 Declined consent to extension of study
- p. 5 Figure 1
20 Declined consent to extension of study
Judge's note: I checked Figure 1 on p.5 and the Discussion on p.9. Only 2 patients were formally lost to follow-up, so 'complete' is defensible in a narrow sense. But Figure 1 puts these patients under 'did not complete follow-up', so the wording overstates. Patients could have more than one reason, so the total is approximate.
Second check: another model agreed
Report a problem with this finding
Wording In Table 1 the 'Primary cause of end-stage renal disease (%)' categories sum to 103.0% (early start) and 103.4% (late start), and no footnote says that categories can overlap.
- p. 6 Table 1, Primary cause of end-stage renal disease (%)
Primary cause of end-stage renal disease (%)
Only the authors can say what is right here.
Judge's note: I read the values on the p.6 image and checked the footnotes on p.7. The footnotes cover only race, BMI and unit conversions; none explains how primary-cause categories could overlap. The race categories in the same table sum to about 100%. Rounding cannot account for a 3-point excess.
Second check: another model agreed
Report a problem with this finding
Typo In the Appendix, "Staff of the Data Management Center Staff" repeats the word 'Staff'.
- p. 11 Appendix
Staff of the Data Management Center Staff
Judge's note: The p.11 page image shows 'Staff of the' followed by italic 'Data Management Center Staff'. The other headings read 'Coordinating Center Staff' and 'Regional Coordinating Centers Staff', so the extra word is a copy-editing slip.
Second check: none (judge only)