Imaging Brain Amyloid in Alzheimer's Disease with Pittsburgh Compound-B
2004 · paper · Biomedicine
Reports the first human study of the PET tracer Pittsburgh Compound-B in 16 mild Alzheimer's patients and 9 controls. Patients showed markedly higher tracer retention in amyloid-rich cortical areas, suggesting it can quantify amyloid deposits in living people.
5 findings
Substance Abstract gives "six older healthy controls (69.5 ± 11 years)" but Table 2 gives OC age 69.0 ± 7.2, and Methods say the six OCs ranged from 59 to 77 years.
- p. 1 Abstract
six older healthy controls (69.5 ± 11 years)
- p. 3 Table 2, OC Age
69.0 ± 7.2
- p. 2 Human Subjects, Healthy volunteers
ranging in age from 59 to 77 years old
- p. 1 Abstract
69.5 11 years
Judge's note: Page 1 abstract image: '69.5 ± 11 years'. Page 3 Table 2, OC column: Age 69.0 ± 7.2. Page 2, Human Subjects: 'Six older controls (OCs) ranging in age from 59 to 77 years old'. All three describe the same six people. For n = 6 within 59–77, the largest possible sample SD is 9.86 (three at each end), so an SD of 11 is impossible. Table 2 is consistent with the stated range.
Second check: another model agreed
Report a problem with this finding
Reference or label Ref 39 gives 'Neurology 1996;96:707–719'; the volume number 96 is implausible for 1996.
- p. 14 Reference 39
Neurology 1996;96:707–719
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Neurology 1996 is volume 46. The internal evidence is enough to show an error: the same list has Neurology 2000;55 (ref 42) and 2001;56 (ref 40), so a 1996 volume of 96 is impossible. The page 14 crop shows '1996;96:707–719'.
Second check: another model agreed
Report a problem with this finding
Reference or label Ref 44 gives 'N Engl J Med 1996;96:752–758'; the volume number 96 repeats the year.
- p. 14 Reference 44
N Engl J Med 1996;96:
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the Reiman et al. 1996 NEJM paper is in volume 334. The page 14 crop shows '1996;96:752–758', the same year-as-volume pattern as ref 39.
Second check: another model agreed
Report a problem with this finding
Reference or label Reference 39 gives the volume as "Neurology 1996;96:707–719"; the volume does not fit the journal's numbering for 1996.
- p. 14
Morris JC, Storandt M, McKeel DW Jr, et al. Cerebral amyloid deposition and diffuse plaque
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: Crossref lists this paper (doi:10.1212/wnl.46.3.707) as volume 46. The page image (crop C033-a) shows "Neurology 1996;96:707–719." The paper's own list also points to a mistake: ref 42 is Neurology 2000;55 and ref 40 is Neurology 2001;56, so a 1996 Neurology volume cannot be 96. The same "1996;96" pattern appears in ref 44. The files contain no erratum.
Second check: another model agreed, code check
Report a problem with this finding
Reference or label Reference 44 gives the volume as "N Engl J Med 1996;96: 752–758"; Crossref lists volume 334.
- p. 14
Reiman EM, Caselli RJ, Yun LS, et al. Preclinical evidence of Alzheimer’s disease in perso
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: Crossref lists this paper (doi:10.1056/nejm199603213341202) as NEJM volume 334. The page image (crop C034-a) shows "N Engl J Med 1996;96: 752–758." Ref 39 has the same suspicious "1996;96", which points to a systematic slip in the reference list. The files contain no erratum.
Second check: another model agreed, code check
Report a problem with this finding
17 minor (typos and the like; most checked by the judge only)
Reference or label Reference 41 prints the co-author as "Morris JC, Price AL.", while reference 40 from the same group prints "Price JL".
- p. 14 Reference 41
Morris JC, Price AL.
- p. 14 Reference 40
Goldman WP, Price JL, Storandt M
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the co-author of Morris JC in J Mol Neurosci 2001;17:101–118 is J.L. Price. The page image (crop C031-a/b, page 14) shows "Morris JC, Price AL." in ref 41 and "Goldman WP, Price JL, Storandt M" in ref 40. The initials do not match between the two entries. The files contain no erratum.
Second check: another model agreed
Report a problem with this finding
Wording Abstract prints "parietal (1.71-fold, p = 0.0002)" but Table 3 gives AD vs HC parietal p = 0.0001 for the same comparison (same 1.71 ratio).
- p. 1 Abstract
parietal (1.71-fold, p = 0.0002)
- p. 9 Table 3, Par column, AD vs HC (p)
0.0001
- p. 1 Abstract
parietal (1.71-fold, p ⴝ 0.0002)
Judge's note: Page 1 image: the abstract reads 'p = 0.0002' with an equals sign. Page 9 image: Table 3, Par column, AD vs HC (p) = 0.0001 and AD:HC ratio = 1.71, so it is the same SUV comparison. Methods (p5) name a single test, Welch two-tailed t. The abstract value matches the Fig 6 '***' threshold (p < 0.0002), so it looks like a threshold printed as an equality. Both values are below the 0.006 Bonferroni threshold, so significance does not change.
Second check: another model agreed
Report a problem with this finding
Wording Abstract prints "temporal (1.52-fold, p = 0.002)" but Table 3 gives AD vs HC temporal p = 0.0011, which does not round to 0.002.
- p. 1 Abstract
temporal (1.52-fold, p = 0.002)
- p. 9 Table 3, Temp column, AD vs HC (p)
0.0011
Judge's note: Page 1 abstract: 'temporal (1.52-fold, p = 0.002)'. Page 9 Table 3, Temp column: p = 0.0011, ratio 1.52, so it is the same comparison. The abstract value matches the Fig 6 '**' threshold (p < 0.002) written with '='. Occipital (0.0017 → 0.002) is consistent. Both values are below 0.006, so significance does not change.
Second check: another model agreed
Report a problem with this finding
Wording Abstract prints "the striatum (1.76-fold, p = 0.0001)" while Table 3 gives p = 0.00004 for the same comparison.
- p. 1 Abstract
the striatum (1.76-fold, p = 0.0001)
- p. 9 Table 3, Striatum column
0.00004
Judge's note: Page 1 abstract uses '=' and gives 0.0001. Page 9 Table 3, Striatum column: 0.00004, ratio 1.76. The abstract value is a conservative upper bound printed as an equality ('<' would be correct). This is minor because the direction and significance are unaffected.
Second check: another model agreed
Report a problem with this finding
Wording Table 1 'MMSE at Diagnosis' mean of 26.6 (SD 2.9) does not equal the mean of the 16 listed values (26.4). It matches only if AD2 (27^d) and AD15 (23^f) are left out, and the table does not say so.
- p. 3 Table 1, MMSE at Diagnosis, Mean row
26.6
Only the authors can say what is right here.
Judge's note: Page 3 Table 1 image: the values sum to 423. 423/16 = 26.44 (SD ≈ 2.83). Leaving out AD2 and AD15, whose footnotes say the scores were not taken at diagnosis, gives 373/14 = 26.64 (SD ≈ 2.87), which matches 26.6 and 2.9. Other column means check out using all rows: duration 69/16 = 4.3, MMSE at study 398/16 = 24.9. The exclusion is plausible but undeclared, so this is an author query.
Second check: another model agreed
Report a problem with this finding
Wording Results say 'The plasma concentration of PIB was studied in the nine HC subjects and 16 AD patients', but Metabolite Analysis describes blood sampling for only 8 HC (4 OC arterialized-venous; 2 of the 3 YC and 2 of the 6 OC radial artery).
- p. 5 Results, Plasma Metabolism of PIB
The plasma concentration of PIB was studied in the nine HC subjects and 16 AD patients.
- p. 4 Metabolite Analysis
Radial artery blood samples were obtained in 2 of the 3 YC subjects, 2 of the 6 OC subjects
Only the authors can say what is right here.
Judge's note: Page 4 Metabolite Analysis: arterialized-venous samples from 13 AD and 4 OC; radial-artery samples from 2 of 3 YC, 2 of 6 OC and 3 of 16 AD. Image Quantitation also says arterial data were available for 7 of 25 subjects (2+2+3). No sampling is described for the third YC, so at most 8 HC are accounted for (fewer if the groups overlap). Page 5 says nine HC. The authors should say whether the third YC was sampled.
Second check: another model agreed
Report a problem with this finding
Wording Results: 'relative differences in rCMRglc ... were in general smaller (eg, 41% in parietal cortex; see Table 4) than ... PIB retention (eg, 94% in frontal cortex; see Table 3)'. Discussion repeats 'approximately 40% in the parietal cortex'. Table 4 gives parietal AD:HC ratio 0.71 (a 29% reduction relative to HC).
- p. 9 Results, Comparison of PIB Retention to Cerebral Metabolic Rate
eg, 41% in parietal cortex
- p. 9 Table 4, Par AD:HC ratio
0.71
- p. 12 Discussion
approximately 40% in the parietal cortex
- p. 9 Results, Comparison of PIB Retention to Cerebral Metabolic Rate
eg, 41% in parietal cortex; see Table 4
Only the authors can say what is right here.
Judge's note: Table 4 on page 9: parietal AD 30.6, HC 43.2, ratio 0.71. The 41% figure is reached only as HC/AD − 1 (AD as baseline). The 94% PIB figure is AD/HC − 1 (HC as baseline; Table 3 ratio 1.94). So the two percentages that the sentence compares use opposite baselines, and the cited Table 4 shows 0.71, not 41%. The ~1.5 SD figure in the page 12 Discussion fits the 12.6 difference. The authors should state which baseline is intended.
Second check: another model agreed
Report a problem with this finding
Wording Several Table 4 correlation p values (eg, Temp r = −0.31, p = 0.187; Striatum r = −0.26, p = 0.323; SWM r = −0.08, p = 0.99; Cb r = −0.19, p = 0.776; n = 24) and r values do not correspond under the declared test (Pearson r, two-tailed).
- p. 9 Table 4, Correlation of rCMRglc to PIB SUV
0.187
- p. 9 Table 4, All subjects SWM
0.99
Only the authors can say what is right here.
Judge's note: Methods (page 5) name the Pearson product moment correlation. I recomputed t = r*sqrt(n-2)/sqrt(1-r^2) with df 22. The Fr, Par and Occ p values (0.037, 0.0001, 0.017) are reproduced, and so is AD-only Par (0.015). Temp gives t≈1.53, two-sided p≈0.14 (printed 0.187). Striatum gives t≈1.26, p≈0.22 (printed 0.323). Pons gives t≈1.37, p≈0.19 (printed 0.266). SWM gives t≈0.38, p≈0.71 (printed 0.99). Cb gives t≈0.91, p≈0.37 (printed 0.776). In the AD-only rows, SWM r = 0.27 with n = 15 gives p≈0.33 (printed 0.438). Rounding r to 2 dp cannot account for these gaps. No adjustment or capping (eg, the repeated 0.99) is described. The authors should say which test and adjustment produced these values.
Second check: another model agreed
Report a problem with this finding
Wording Discussion says the absolute amount of PIB retained in the frontal cortex of AD patients 'was greater than 90% higher than that retained in control frontal cortex or cerebellum of either controls or AD patients'; Table 3 gives AD frontal 1.56 vs AD cerebellum 0.86 (about 81% higher).
- p. 10 Discussion, para 2
greater than 90% higher than that retained in control frontal cortex or cerebellum of either controls or AD patients
- p. 9 Table 3, AD mean Fr and Cb
1.56
- p. 9 Table 3, Fr and Cb columns
0.86
Only the authors can say what is right here.
Judge's note: Table 3 on page 9: AD Fr 1.56, AD Cb 0.86, HC Fr 0.80, HC Cb 0.74. The >90% claim holds against HC frontal cortex and HC cerebellum, but not against AD cerebellum using the 40–60 min SUVs. The sentence cites the time-activity curves, so a particular time point might give a different ratio; only the authors can say.
Second check: another model agreed
Report a problem with this finding
Typo Subject–verb disagreement: "The exact form of the amyloid target of BTA-1 binding have not yet been identified with certainty" (should be 'has').
- p. 1 Introduction
exact form of the amyloid target of BTA-1 binding
- p. 1 Introduction
The exact form of the amyloid target of BTA-1 binding have not yet been identified
Judge's note: Page 1 image, Introduction: the sentence runs across the column break and reads as quoted. The singular subject 'form' takes 'have'.
Second check: none (judge only)
Report a problem with this finding
Typo Redundant noun: "recruitment and evaluation of AD patients subjects".
- p. 2 Subjects and Methods, first paragraph
recruitment and evaluation of AD patients subjects
Judge's note: Page 2 image, Subjects and Methods, first paragraph: reads 'AD patients subjects'. One of the two nouns is redundant.
Second check: none (judge only)
Report a problem with this finding
Typo "National Institute of Neurological and Communication Disorders Alzheimer's Disease and Related Disorders Association criteria" leaves out 'and Stroke' (and has 'Communication'), so it does not match the NINCDS-ADRDA acronym in ref 14.
- p. 3 Alzheimer's disease patients paragraph
National Insti-tute of Neurological and Communication Disorders Alzhei-mer's Disease and Related Disorders Association criteria
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: NINCDS stands for National Institute of Neurological and Communicative Disorders and Stroke. Page 3 image shows the expansion as quoted. Ref 14 (McKhann et al) uses 'NINCDS-ADRDA', and the expansion has no word for the 'S'.
Second check: none (judge only)
Report a problem with this finding
Typo Methods subsection heading is misspelled 'Statstical Analyses' (should be 'Statistical Analyses').
- p. 5 Methods subsection heading
Statstical Analyses
Judge's note: The misspelling is visible in the heading at the top of the left column on the page 5 image.
Second check: none (judge only)
Report a problem with this finding
Typo Disclosure has 'Dr Langstrom', while the author list and ref 13 spell the name 'Långström'.
- p. 12 Disclosure
Dr Langstrom
Judge's note: Page 12 image shows 'Dr Langstrom' without diacritics. The author line on page 1 and ref 13 use Långström (as the text layer shows). This is a cosmetic inconsistency.
Second check: none (judge only)
Report a problem with this finding
Typo Funding note: 'the NIH (National Institute of Aging), AG01039, AG20226, W.K.; AG18402 C.M.),' has an unmatched closing parenthesis after 'C.M.'.
- p. 12 Funding note
the NIH (National Institute of Aging), AG01039, AG20226, W.K.; AG18402 C.M.),
Judge's note: On the page 12 image, the parenthesis opened at 'NIH (' closes after 'Aging'. The closing parenthesis after 'C.M.' has no opening partner. The parallel entries elsewhere in the note suggest the intended form is '(... AG18402, C.M.)'.
Second check: none (judge only)
Report a problem with this finding
Typo Ref 21 reads 'Logan J, Fowler AH, Volkow ND, et al.' whereas ref 20 reads 'Logan J, Fowler JS, Volkow ND, et al.'.
- p. 13 Reference 21
Logan J, Fowler AH, Volkow ND
- p. 13 Reference 20
Logan J, Fowler JS, Volkow ND
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the ref 21 co-author is J.S. Fowler. On the page 13 crop, both companion references have the same author sequence (Logan, Fowler, Volkow) in J Cereb Blood Flow Metab, but the initials differ: 'AH' in ref 21, 'JS' in ref 20.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 49 prints "Berg L, McKeel DWJ, Miller JP", while reference 39 gives the same author as "McKeel DW Jr".
- p. 14 Reference 49
Berg L, McKeel DWJ, Miller JP
Judge's note: The page image (crop C032-a) shows "McKeel DWJ" in ref 49. Ref 39 on the same page has "McKeel DW Jr". The suffix has been run into the initials, so the two entries format the same author differently. This is a cosmetic formatting slip.
Second check: none (judge only)