Cardiac-Resynchronization Therapy with or without an Implantable Defibrillator in Advanced Chronic Heart Failure
ACC Current Journal Review, 2004 · paper · Clinical trials
A randomized trial of 1520 patients with advanced heart failure and prolonged QRS interval, comparing drugs alone with resynchronization pacing, with or without a defibrillator. Both device types reduced death or hospitalization; only the pacemaker–defibrillator significantly reduced mortality.
2 findings
Reference or label The Figure 3B y-axis ticks read 2, 3, 0, −1, −2 from top to bottom; the tick between 2 and 0 should be 1.
- p. 9 Figure 3, Panel B y-axis
Change in Diastolic Blood Pressure from Baseline (mm Hg)
Judge's note: The p. 9 crop shows equally spaced ticks labelled '2', '3', '0', '−1', '−2' on the diastolic blood pressure panel. This is a mislabelled axis tick.
Second check: another model agreed
Report a problem with this finding
Reference or label Reference 20 prints "Lan KKG, DeMets DL. Discrete sequential boundaries for clinical trials. Biometrika 1983;70:649-53."; the article's pages are 659-63.
- p. 11 Reference 20
Biometrika 1983;70:649-53.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: Lan & DeMets, Biometrika 1983;70(3):659-663. The page image (p11 reference list, crop C019-a) shows '70:649-53' as quoted. This is a citation page-range error.
Second check: another model agreed
Report a problem with this finding
12 minor (typos and the like; most checked by the judge only)
Wording The pacemaker heart-failure composite is reported two ways: 'reduced by 34 percent in the pacemaker group (P<0.002)' in the abstract, but 'P=0.002' in the Results (p. 7) and Figure 1D ('237 events, P=0.002'). The interval '(hazard ratio, 0.66; 95 percent confidence interval, 0.53 to 0.87...)' is not centred on 0.66 on the log scale.
- p. 1 Abstract, Results
(P<0.002)
- p. 7 Results, death from or hospitalization for heart failure
interval, 0.53 to 0.87; P=0.002
- p. 4 Figure 1D label
237 events, P=0.002
- p. 7 Death from or hospitalization for heart failure paragraph
hazard ratio, 0.66; 95 percent confidence interval, 0.53 to 0.87
Only the authors can say what is right here.
Judge's note: Abstract (p. 1 image): 'P<0.002'. Results (p. 7 image) and Fig. 1D (p. 4 image): 'P=0.002' for the same comparison. One quantity is printed two ways for evidently the same analysis, so this is confirmed as a wording inconsistency. On the interval: the geometric midpoint of 0.53 and 0.87 is about 0.679, against a printed HR of 0.66. Every other HR/CI pair in the paper is log-centred to within 0.01: 0.81 (0.69-0.96), 0.80 (0.68-0.95), 0.76 (0.58-1.01), 0.75 (0.63-0.90), 0.72 (0.60-0.86), 0.60 (0.49-0.75), 0.50 (0.29-0.88), 0.73 (0.52-1.04), 0.91 (0.55-1.49), 0.72 (0.51-1.01). This interval is also wider on the log scale than the pacemaker-defibrillator interval, which rests on fewer events. Question for the authors: is 0.53 to 0.87 the interval for the HR of 0.66 from the same unadjusted Cox model, or should one bound read differently? I make no claim about which number is wrong.
Second check: another model agreed
Report a problem with this finding
Wording In the Figure 2 pacemaker column, two subgroup pairs sum to 616 against n=617: cardiomyopathy (Ischemic 331 + Nonischemic 285) and bundle-branch block (Left 426 + Other 190).
- p. 6 Figure 2, Pacemaker column, Cardiomyopathy and Bundle branch block rows
331
- p. 6 Figure 2, Pacemaker column, Bundle branch block rows
426
- p. 6 Figure 2, Cardiomyopathy, pacemaker column
Ischemic 181 331 325 / Nonischemic 127 285 270
Only the authors can say what is right here.
Judge's note: From the upright p. 6 image, every other pacemaker pair sums to 617 (age, sex, NYHA, LVEF, QRS, heart rate, BP, drugs), and the other two arms sum to 308 and 595 for these variables. The caption notes missing values only for LVEDD. One patient may be missing etiology or ECG data, but the paper does not say so.
Second check: another model agreed
Report a problem with this finding
Wording The text says subgroup hazard ratios for the primary end point were 'consistently below 1.0', but in Figure 2 the pacemaker point estimate for QRS '≤147 msec' sits on the 1.0 line.
- p. 7 Results, subgroup analyses
teristics that were consistently below 1.0, indicating
- p. 6 Figure 2, pacemaker panel, QRS ≤147 msec row
≤147 msec
Only the authors can say what is right here.
Judge's note: In the upright p. 6 image, the square for ≤147 msec in the pacemaker primary-end-point panel is centred on the vertical 1.0 reference line. All other primary-end-point estimates in both device panels lie left of 1.0. The text claim overstates for that one subgroup unless the estimate is just below 1.00.
Second check: another model agreed
Report a problem with this finding
Wording Discussion states "our data demonstrate that adding a defibrillator to cardiac-resynchronization therapy significantly reduces the risk of death", but every mortality P value reported compares a device group with pharmacologic therapy (pacemaker P=0.059; pacemaker–defibrillator P=0.003). No pacemaker–defibrillator vs pacemaker test is shown.
- p. 10 Discussion, first paragraph of column 1
adding a defibrillator to cardiac-resynchronization therapy significantly reduces the risk of death
- p. 9 Discussion, column 2
resynchronization therapy incrementally increased the survival benefit
- p. 7 Results, secondary end point
hazard ratio, 0.64; 95 percent confidence interval, 0.48 to 0.86; P=0.003
Only the authors can say what is right here.
Judge's note: Checked p7 results (HR 0.76, P=0.059 for pacemaker; HR 0.64, 0.48 to 0.86, P=0.003 for pacemaker–defibrillator, both vs pharmacologic therapy). The p9 discussion does qualify the 36 percent reduction as 'as compared with optimal pharmacologic therapy', but the p10 sentence attributes significance to the addition of the defibrillator. The methods (power statement, lines ~794-798) and the table footnotes describe comparisons with pharmacologic therapy only. I found no direct test between the two device groups in the text. This is a wording overstatement and only the authors can clarify it.
Second check: another model agreed
Report a problem with this finding
Wording Table 2 row header 'Increase in quality of life (%)†' sits over negative mean changes (e.g., −9±21, −24±27), while the footnote defines a Minnesota score (total 105) where higher means worse.
- p. 8 Table 2, quality-of-life row header
Increase in quality of life (%)†
Only the authors can say what is right here.
Judge's note: The p. 8 crop shows the label and negative values. The footnote says 'higher scores indicating a poor quality of life'. Negative values under a label of 'Increase', with '%' as the unit for what appears to be a score change, leave the direction and unit ambiguous. Only the authors can say whether these are points or percent.
Second check: another model agreed
Report a problem with this finding
Typo Affiliation reads 'Moffit Hos-pital, University of California, San Francisco'.
- p. 1 Author affiliations
Moffit Hos-pital, University of California, San Francisco
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the UCSF hospital is spelled 'Moffitt'. The page 1 image shows 'Moffit Hos-pital'.
Second check: none (judge only)
Report a problem with this finding
Typo The text says 'synchronizes the activation of the intra-ventricular septum and left ventricular free wall'; the septum between the ventricles is the interventricular septum.
- p. 2 Introduction, paragraph 1
synchronizes the activation of the intra-ventricular septum
- p. 2 Introduction, paragraph 1
synchronizes the activation of the intra-ventricular septum and left ventricular free wall
Judge's note: The page 2 image shows 'intra-' and 'ventricular septum' split across a line break, i.e. 'intraventricular septum'. The structure between the ventricles is the interventricular septum. The wording looks carried over from 'intraventricular conduction delays' on the line above. The meaning is still recoverable, so S4.
Second check: none (judge only)
Report a problem with this finding
Typo A Figure 2 subgroup label reads 'Heart ratio' for the ≤72 / >72 beats/min split; it should be 'Heart rate'.
- p. 6 Figure 2, subgroup variable list
Heart ratio
Judge's note: The upright p. 6 image shows 'Heart ratio' above '≤72 beats/min' and '>72 beats/min'. Table 1 lists 'Heart rate (beats/min)' with a median of 72.
Second check: none (judge only)
Report a problem with this finding
Typo In the Appendix, 'Willis-Knighton Hospital; Shreveport, La.' uses a semicolon where the list's style puts a comma between institution and city.
- p. 10 Appendix, investigator list
Willis-Knighton Hospital; Shreveport, La.
Judge's note: The p. 10 crop confirms the semicolon. Every other entry uses a comma between institution and city.
Second check: none (judge only)
Report a problem with this finding
Typo Appendix prints "A. Ciuffo, Sentera Norfolk General Hospital, Norfolk, Va."; the institution is normally spelled Sentara.
- p. 10 Appendix
Sentera Norfolk General Hospital
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the Norfolk, Va., hospital is spelled 'Sentara Norfolk General Hospital'. The page image (p10 appendix, crop C016-a) shows 'Sentera' exactly as quoted. Cosmetic misspelling of a proper name.
Second check: none (judge only)
Report a problem with this finding
Typo Appendix prints "Deborah Heart and Lung Center, Brown Mills, N.J."; the place name is normally Browns Mills.
- p. 10 Appendix
Heart and Lung Center, Brown Mills, N.J.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: Deborah Heart and Lung Center is in Browns Mills, New Jersey. The page image (p10, crop C017-a) shows 'Brown Mills, N.J.' as quoted. Cosmetic misspelling.
Second check: none (judge only)
Report a problem with this finding
Typo Appendix prints "J. Schmedtje, Carillion Medical Center, Roanoke, Va."; the Roanoke system is spelled Carilion.
- p. 11 Appendix
Carillion Medical Center, Roanoke, Va.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the Roanoke, Va., health system is spelled 'Carilion'. The page image (p11 appendix, crop C020-a) shows 'Carillion' as quoted. Cosmetic misspelling.
Second check: none (judge only)