To the Editor: We read with concern the analysis by Davidsonet al. (Sept. 26 issue).1 They use the pooled placebo groupsfrom two sepsis trials, divided into two groups — patientswho either were receiving heparin at the time they entered thetrial or began heparin therapy during the trial and those whonever received heparin.2,3 Patients who began receiving heparintreatment during the trials were moved from the nonheparin groupto the heparin group. Not only was the intention-to-treat populationtherefore unknown at trial entry, but also patients who survivedlonger had a greater chance of receiving heparin. This uncontrolled,unidirectional movement of patients receiving new heparin treatmentduring the trial preferentially selects for those who surviveuntil treatment begins, creating a fundamental flaw in the analysis.The observed benefit associated with the use of heparin maybe entirely due to the movement of patients from the nonheparingroup to the heparin group. Thus, the analysis by Davidson etal. does not provide evidence of a benefit from heparin in severesepsis.
Ross Freebairn, M.B., Ch.B. Hawke's Bay Hospital Hastings 4201, New Zealand
Sarah Ramsay, M.B., B.S. Charles Gomersall, M.B., B.S. Chinese University of Hong Kong Hong Kong, China
References
Davidson BL, Geerts WH, Lensing AWA. Low-dose heparin for severe sepsis. N Engl J Med 2002;347:1036-1037. [Free Full Text]
Bernard GR, Vincent J-L, Laterre P-F, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001;344:699-709. [Free Full Text]
Warren BL, Eid A, Singer P, et al. Caring for the critically ill patient: high-dose antithrombin III in severe sepsis: a randomized controlled trial. JAMA 2001;286:1869-1878. [Erratum, JAMA 2002;287:192.] [Free Full Text]
The authors reply: We disagree with the explanation proposed by Freebairn et al. to account for our findings. We wish toclarify that the report on the antithrombin III study1 and theanalysis by the Food and Drug Administration of the trial ofactivated protein C,2 the sources of our data, explicitly defineheparin recipients as those receiving heparin from day 1 today 4 — the same period during which the study drugs wereadministered. Therefore, longer survival is not required to"enter" the heparin group, and such a bias could not reasonablyexplain our results.
We reiterate that our findings are hypothesis-generating. Giventhe strong suggestion of the efficacy as well as the safetyand low expense of heparin, a prospective trial is warrantedto compare low-dose heparin, low-molecular-weight heparin, orboth with placebo and activated protein C in order to assessthe relative contributions of these agents to the survival ofpatients with sepsis.
We apologize for the oversight but wish to disclose that Drs.Davidson and Geerts have served as clinical investigators, consultants,or both for one or more of the following manufacturers of low-molecular-weightheparin: Aventis, Sanofi, and Pharmacia.
Bruce L. Davidson, M.D., M.P.H. Swedish Medical Center Seattle, WA 98122 brucedavidson{at}pobox.com
William H. Geerts, M.D. Sunnybrook and Women's College HealthSciences Centre Toronto, ON M5S 1B2, Canada
Anthonie W.A. Lensing, M.D., Ph.D. Academic Medical Center 1105 AZ Amsterdam, the Netherlands
References
Warren BL, Eid A, Singer P, et al. Caring for the critically ill patient: high-dose antithrombin III in severe sepsis: a randomized controlled trial. JAMA 2001;286:1869-1878. [Erratum, JAMA 2002;287:192.] [Free Full Text]
Anti-Infective Advisory Committee. FDA briefing document: drotrecogin alfa (activated) [recombinant human activated protein C (rhAPC)]: Xigris. BLA no. 125029/0. Rockville, Md.: Food and Drug Administration, September 2001. (Accessed February 12, 2003, at http://www.fda.gov/ohrms/dockets/ac/01/briefing/3797b1_02_FDAbriefing.pdf.)
Correction
to
Davidson et al., 
Previous
