Spinal Anesthesia or General Anesthesia for Hip Surgery in Older Adults
2021 · paper · Clinical trials
A randomized trial of spinal versus general anesthesia in 1600 older adults having hip-fracture surgery. Death or inability to walk at 60 days was similar (18.5% vs 18.0%), as was delirium.
Checked: PDF from web.archive.org, fetched 5 Oct 2026. Page numbers are those of that PDF.
1 finding
Substance Figure 1, general anesthesia arm: "71 Did not have data available for primary outcome" lists "1 Withdrew before interview", "44 Were unable to be contacted", "16 Were contacted but walking status was unknown", which sum to 61, not 71.
- p. 5 Figure 1, general anesthesia arm follow-up box
71 Did not have data available for primary outcome
- p. 5 Figure 1, general anesthesia arm follow-up box
1 Withdrew before interview
- p. 5 Figure 1, general anesthesia arm allocation box
7 Withdrew before surgery or did not have data on anesthesia type received
- p. 5 Results text, page 2029
Seven patients who had been assigned to gen-eral anesthesia (0.9%) withdrew consent before surgery
Only the authors can say what is right here.
Judge's note: Checked Figure 1 on the p5 image. The subcounts read 1, 44 and 16. The total of 71 holds up elsewhere: 805−734 = 71, and the lower box gives 71 missing. The spinal arm adds up (13+56+14 = 83). The text and the figure both say 7 general-anesthesia patients withdrew before surgery with no outcome data collected, which does not fit '1 Withdrew before interview'. One subcount is probably misprinted (11 would fit), but only the authors can say which.
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: "The primary analysis and analyses of all secondary ... outcomes included patients in the modified intention-to-treat population for whom complete data were available", with IPW only "To assess the effect of missing data"; yet Table 2 (footnote *: "Results shown for the primary outcome comparison reflect inverse-probability weighting") and the Abstract ("for whom data were available ... 95% CI, 0.84 to 1.27") present the IPW result as the primary result.
- p. 4 Statistical Analysis
included patients in the modified inten-tion-to-treat population for whom complete data were available for the relevant outcomes
- p. 4 Statistical Analysis
To assess the effect of missing data on the findings for the primary outcome, we performed an inverse-probability–weighted analysis
- p. 7 Table 2 footnote *
Results shown for the primary outcome comparison reflect inverse-probability weighting to account for missing outcome data
- p. 1 Abstract, Results
relative risk, 1.03; 95% confidence interval [CI], 0.84 to 1.27
Only the authors can say what is right here.
Judge's note: Checked on the page images. The Methods text runs from p3 to p4: the complete-case analysis is primary and IPW is a sensitivity analysis. Results text on p5: complete case 0.83 to 1.28, IPW 0.84 to 1.27. The Table 2 footnote on p7 says outright that the primary row is IPW. The Abstract on p1 gives 0.84 to 1.27 and does not say it is IPW, while calling the population the patients 'for whom data were available'. The two CIs are not a numeric conflict, because they come from different analyses. What is open is which analysis is primary, so this is a question for the authors.
Second check: another model agreed
Report a problem with this finding
Wording Results text "and 7.6% were Black" uses all 1600 randomized patients as denominator (122/1600), whereas Table 1 reports race among those with race data (55/762 and 67/774, pooled 122/1536 = 7.9%); denominator not stated.
- p. 4 Results, Patients and Treatment
and 7.6% were Black
- p. 6 Table 1, Race
55/762 (7.2)
- p. 6 Table 1, Race
67/774 (8.7)
- p. 4 Results, Patients and Treatment
33.0% were men, and 7.6% were Black
Judge's note: Checked Table 1 (p6) and the text (p4) on the images. 7.6% can only be reached with 1600 as the denominator. The same text's '33.0% were men' also uses 1600 (528/1600), so the choice is applied consistently. It is still not stated, and it differs from the Table 1 convention. The Discussion's 'approximately 8%' fits either figure. This is a minor point about denominators, not a wrong number.
Second check: another model agreed
Report a problem with this finding
Wording Outcomes text: "132 of 712 patients (18.5%) who received spinal anesthesia and in 132 of 733 patients (18.0%) who received general anesthesia" describes the modified intention-to-treat groups, which are defined by assignment, as groups defined by anesthesia received.
- p. 5 Results, Outcomes, first paragraph
132 of 712 patients (18.5%) who received spinal anesthesia and in 132 of 733 patients (18.0%) who received general anesthesia
- p. 4 Methods, Statistical Analysis
Patients were included in the analysis according to their original treatment assignment.
Judge's note: Checked p4 (Methods: 'Patients were included in the analysis according to their original treatment assignment') and p5. 119 of the 795 spinal-assigned patients received general anesthesia. Table 2, the Abstract and the delirium paragraph say 'assigned to'. The wording here should say 'assigned to'.
Second check: another model agreed
Report a problem with this finding
Typo Table 4 footnote ¶: "The 12-item World Health Organization Disability Schedule 2.0 (WHODAS 2.0)" omits 'Assessment' (Methods and ref 22: 'Disability Assessment Schedule').
- p. 9 Table 4 footnote ¶
The 12-item World Health Organization Disability Schedule 2.0 (WHODAS 2.0)
- p. 3 Outcomes
12-item World Health Organization Disability Assessment Schedule 2.0
Judge's note: The p9 image shows the word missing. The Methods on p3 and ref 22 on p11 have 'Disability Assessment Schedule'.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 13: "Horlocker TT, Vandermeuelen E, Kopp SL" — author surname likely 'Vandermeulen'.
- p. 11 Reference 13
Horlocker TT, Vandermeuelen E, Kopp SL
- p. 11 Reference 13
Vandermeuelen E
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the co-author of the ASRA 4th-edition guidelines spells the name 'Vandermeulen'. The p11 image shows 'Vandermeuelen'.
Second check: none (judge only)