RSS Feeds
Albumin Meta-Analysis
- Mahlon M. Wilkes, PhD; and
- Roberta J. Navickis, PhD
- Hygeia Associates; Grass Valley, CA 95949 Note: The authors also acknowledged in their cover letter that they had served as consultants to the Plasma Protein Products Therapeutics Association. They further disclosed that they otherwise held no financial interest of any kind in the Plasma Protein Therapeutics Association, any albumin products, or products competing with albumin.—The Editor
The Editors welcome submissions for possible publication in the Letters section. Authors of letters should:
•Include no more than 300 words of text, three authors, and five references
•Type with double-spacing
•Send three copies of the letter, an authors' form signed by all authors, and a cover letter describing any conflicts of interest related to the contents of the letter.
Letters commenting on an Annals article will be considered if they are received within 6 weeks of the time the article was published. Only some of the letters received can be published. Published letters are edited and may be shortened; tables and figures are included only selectively. Authors will be notified that the letter has been received. If the letter is selected for publication, the author will be notified about 3 weeks before the publication date. Unpublished letters cannot be returned.
Annals welcomes electronically submitted letters.
IN RESPONSE:
Gibney asserts that our meta-analysis “fails to demonstrate indications for colloid” and was biased in favor of albumin therapy because of industry funding. Our meta-analysis addressed albumin safety broadly, in the context of a previous meta-analysis suggesting excess albumin-associated mortality (1). We did not try to assess the efficacy of albumin therapy for particular indications. Optimally, such an assessment would encompass additional clinically relevant end points, such as morbidity and length of stay, rather than solely mortality, which is a comparatively insensitive end point in large populations of acutely ill patients. In several randomized trials, significant differences between albumin and control groups could be demonstrated with respect to morbidity or length of stay but not mortality (2–5).
We detected no overall effect of albumin on mortality. This observation does not support withholding albumin because of safety concerns. We also found that relative risk for death was substantially and consistently lower in trials of higher quality, as judged by blinding, the presence of mortality as a study end point, absence of crossover, and larger trial size. The results in both blinded trials and larger trials, as well as trials sharing two or more characteristics indicative of higher quality, suggested that albumin may reduce mortality. This possibility merits further study in future well-designed trials.
We agree that study selection bias can affect meta-analyses. In the previous meta-analysis (1), one third of the eligible patients were excluded, and among these patients, mortality was lower in the albumin group. Thus, the exclusion rules applied in that meta-analysis biased the results in favor of control. Our method, however, was to include “all comers.” This approach was designed to avoid bias arising from exclusion rules. The Plasma Protein Therapeutics Association and the American Red Cross provided funding but played no role in the conduct of our study, the analysis of results, or reporting. The decision to support our work was made long before the results of the meta-analysis were known, and the funding organizations did not retain any right to block publication depending on our findings. Our data were derived from published randomized trials, and we described our methods in sufficient detail that our findings can be easily replicated by others. In view of these facts, we see no sound basis for a generic argument that industry funding per se would have biased our results.
Mahlon M. Wilkes, PhD
Roberta J. Navickis, PhD
Hygeia Associates
Grass Valley, CA 95949
The Editors welcome submissions for possible publication in the Letters section. Authors of letters should:
•Include no more than 300 words of text, three authors, and five references
•Type with double-spacing
•Send three copies of the letter, an authors' form signed by all authors, and a cover letter describing any conflicts of interest related to the contents of the letter.
Letters commenting on an Annals article will be considered if they are received within 6 weeks of the time the article was published. Only some of the letters received can be published. Published letters are edited and may be shortened; tables and figures are included only selectively. Authors will be notified that the letter has been received. If the letter is selected for publication, the author will be notified about 3 weeks before the publication date. Unpublished letters cannot be returned.
Annals welcomes electronically submitted letters.
- Copyright ©2004 by the American College of Physicians
