A Comparison of Laparoscopically Assisted and Open Colectomy for Colon Cancer
New England Journal of Medicine, 2004 · paper · Clinical trials
A noninferiority trial of 872 patients with colon cancer, comparing laparoscopically assisted with open colectomy. Three-year recurrence was 16% versus 18% and survival was similar; recovery was faster with laparoscopy, suggesting it is an acceptable alternative.
1 finding
Substance Text gives concurrent resection of other organs as "(63 vs. 34 patients, P=0.001)"; Table 2 gives "Other organs resected — no. (%)" as "62 (14)" and "33 (8)" with the same P value, and the footnote's organ counts sum to 62 and 33.
- p. 4 Results, Surgery, second paragraph
(63 vs. 34 patients, P=0.001)
- p. 6 Table 2, row 'Other organs resected — no. (%)'
62 (14)
- p. 6 Table 2, row 'Other organs resected — no. (%)', laparoscopic column
33 (8)
- p. 6 Table 2 footnote *
Other organs resected included the gallbladder (10 patients in each group)
Judge's note: Read the Surgery paragraph on page 4 and Table 2 with its footnote on page 6. Same groups (N=428 and N=435), same variable, same P=0.001. The footnote categories sum to the table's counts, and the printed percentages fit 62/428 and 33/435. A patient with two organs resected would make the footnote sum larger than the patient count, not smaller, so no counting rule explains the text's higher figures. No explanation found in the methods or elsewhere. The text is one higher than the table in each group.
Second check: another model agreed
Report a problem with this finding
4 minor (typos and the like; most checked by the judge only)
Wording Text reads "rates and severity of postoperative complications at discharge (P=0.98) and at 60 days (P=0.73)"; in Table 2, 0.98 is the P value for "Postoperative (before discharge)" complications and 0.73 is the P value for "Grade of postoperative complications" (denominators 80 and 81), and no 60-day complication data are shown.
- p. 4 Results, Recovery and complications
operative complications at discharge (P=0.98) and at 60 days (P=0.73)
- p. 6 Table 2, row 'Grade of postoperative complications — no./total no. (%)'
0.73
- p. 6 Table 2, row 'Postoperative (before discharge)'
0.98
- p. 6 Table 2, Grade of postoperative complications — no./total no. (%)
44/80 (55)
Only the authors can say what is right here.
Judge's note: Read the Recovery and complications paragraph on page 4 and Table 2 on page 6. The two P values in the text match the table's rate (0.98) and grade (0.73) rows, both of which concern complications before discharge (grade denominators 80 and 81 equal the before-discharge counts). The methods say complications were assessed at discharge and at 2 and 18 months, and the abstract says complications "at discharge and 60 days" were very similar, but no 60-day counts or P value appear in the table or text. Either the P values belong to rate and severity and the parentheses are misplaced, or a separate 60-day test also gave 0.73 and the severity P value is not in the text. Only the authors can say which.
Second check: another model agreed
Report a problem with this finding
Wording Disease-free survival is reported with "117 events in the open-colectomy group and 118 events in the laparoscopic-surgery group", while recurrences ("84 in the open-colectomy group and 76 in the laparoscopic-surgery group") plus deaths before recurrence ("34 in the open-colectomy group and 43 in the laparoscopic-surgery group") give 118 and 119.
- p. 5 Results, Survival and recurrence, right column
117 events in the open-colectomy group and 118 events in the laparoscopic-surgery group
- p. 4 Results, Survival and recurrence
84 in the open-colectomy group and 76 in the laparoscopic-surgery group
- p. 4 Results, Survival and recurrence
34 in the open-colectomy group and 43 in the laparoscopic-surgery group
- p. 5 Survival and recurrence
117 events in the open-
Judge's note: Read the Survival and recurrence text on pages 4 and 5 and the Statistical analysis section on page 3. The sums are one higher than the printed event counts in each group. A likely explanation exists but is not stated: patients with benign pathological conditions were excluded from the disease-free survival analysis (methods, page 3), and the paper does not say whether the 186 deaths and the 77 deaths before recurrence are counted among all 863 patients or among the 810 with cancer. One death in a benign-pathology patient per group would account for the difference. Because the paper gives the exclusion but does not tie it to these counts, this is kept as a minor consistency point, not an error.
Second check: another model agreed
Report a problem with this finding
Wording Discussion states that "the length of bowel and mesentery resected, and the bowel margins — did not differ significantly", but the results and Table 2 report only proximal and distal margins, the share with margins under 5 cm and the median number of lymph nodes; no length of bowel or mesentery resected is reported.
- p. 7 Discussion, first full paragraph under Figure 2
the length of bowel and mesentery resected, and the bowel margins — did not differ significantly
- p. 4 Results, Surgery, last paragraph
The extent of resection was similar in both groups; bowel margins were less than 5 cm in 6 percent
- p. 6 Table 2
Proximal margin — cm
Judge's note: Read the discussion paragraph on page 7, the Surgery section on page 4 and all rows of Table 2 on page 6. The results say only "The extent of resection was similar in both groups" and give margins (P=0.38, 0.09, 0.52) and a median of 12 lymph nodes in each group with no P value. No length of bowel or of mesentery appears in text, table or figure, and there is no "data not shown" here. The discussion claim for those two measures therefore has no reported data behind it. Minor: an unsupported statement, not a contradiction.
Second check: another model agreed
Report a problem with this finding
Wording Figure 2, No. at Risk, open colectomy at year 1: panel A (all stages) prints "345", but panels B, C and D (stages I, II, III) print 104, 135 and 107, which sum to 346.
- p. 7 Figure 2A, No. at Risk, open colectomy, year 1
345
- p. 7 Figure 2B, No. at Risk, open colectomy
112 104 97 85 66 39
- p. 7 Figure 2C, No. at Risk, open colectomy
146 135 112 93 69 46
- p. 7 Figure 2D, No. at Risk, open colectomy
121 107 80 62 42 24
Only the authors can say what is right here.
Judge's note: Read the at-risk rows on the page 7 image and on an enlarged crop of the figure. Panel A open colectomy: 395 345 289 240 177 109. Panels B, C, D: 112 104 97 85 66 39; 146 135 112 93 69 46; 121 107 80 62 42 24. The all-stage group contains the three stage groups (plus stage IV and unknown), so its number at risk cannot be below their sum. At year 0 the sum is 379, and 16 stage IV patients bring it to 395. At years 2 to 5 the sums equal panel A exactly (289, 240, 177, 109). In the laparoscopic row the sums never exceed panel A (365 against 368 at year 1). Only the open-colectomy year 1 column is impossible, by one. One of the four printed numbers is off; the files do not show which. No effect on the curves or conclusions.
Second check: another model agreed