LETTERS Corticosteroids for Patients With Septic Shock To the Editor: Dr Annane and colleagues 1 found that corti- costeroids reduced the risk of mortality in septic shock only among patients who did not respond to a corticotropin stimu- lation test. Some of these “nonresponders” had a baseline cor- tisol of less than 34 μg/dL with an increase of less than 9 μg/ dL, and thus many of those may have had absolute adrenal deficiency. Such patients have a high risk of mortality 2 and it may be inappropriate to randomize them to receive placebo. It would also be of interest to know how many of the nonre- sponders in this study had absolute adrenal deficiency and the effect of removing such patients from the analysis. Julian M. Brown, MB, ChB Department of Anaesthesia Frenchay Hospital Bristol, England 1. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288:862-871. 2. Soni A, Pepper GM, Wyrwinski PM, et al. Adrenal insufficiency occurring dur- ing septic shock: incidence, outcome, and relationship to peripheral cytokine lev- els. Am J Med. 1995;98:266-271. To the Editor: Dr Annane and colleagues 1 excluded patients who received etomidate before intubation because this agent could interfere with cortisol response to corticotropin. How- ever, the authors do not provide the number of patients ex- cluded on this basis. Furthermore, this exclusion criterion was not implemented until halfway through the data collection phase. It would be of interest to know how many included in this study received etomidate prior to that. It also would be of interest to know the effect of removing these patients from the data analysis. This would help answer the question of whether patients with septic shock who have received etomidate should be routinely treated with exogenous corticosteroids. Christina L. Schenarts, MD Juan A. March, MD Department of Emergency Medicine East Carolina University Greenville, NC 1. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288:862-871. To the Editor: Dr Annane and colleagues 1 claim that cortico- tropin nonresponders with septic shock who were treated with corticosteroids had a significantly decreased mortality. This claim is based on an adjusted Cox model that corrects for baseline cortisol, cortisol response to tetracosactin, McCabe classifica- tion, Logistic Organ Dysfunction score, arterial lactate con- centration, and PAO 2 /Fio 2 . The authors were able to show a mod- est decrease in mortality only after correcting for these factors. The  2 tests on the 28-day mortality for the study group as a whole and without adjustment for covariates did not reveal significant differences, however. Thus, the authors could have concluded that they failed to demonstrate a survival advan- tage. Randomization should have evened out differences be- tween the placebo and treatment groups, but the authors chose to factor in many confounders and as a result claim a survival advantage for nonresponders. In a pragmatic trial such as this, the most important result is the simple one—the uncorrected effect on overall mortality. Simple tests of proportionality dem- onstrate that the raw results are not statistically significant. Julian Millo, BSc, MRCP, FRCA, DipICM Nuffield Department of Anaesthetics John Radcliffe Hospital Oxford, England 1. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288:862-871. To the Editor: I am concerned that Dr Annane and col- leagues 1 studied a highly selected subgroup of the entire population of septic patients. Of the more than half million patients estimated to have severe sepsis in the United States annually, 2 only a small fraction would have vasopressor- and ventilator-dependent septic shock with documented relative adrenal insufficiency within 1 to 8 hours of the onset of septic shock. Remarkably, only 24% of the patients in this study were bacteremic, yet they experienced an exceedingly high mortality rate of 55% to 61%. The majority of patients had a community-acquired infection with a nonfatal underlying dis- ease state. Such patients should receive optimal supportive care and should have reversible disease physiology. It remains unclear whether corticosteroids benefit a more heterogeneous population with serious underlying diseases or nosocomial infections. Steven M. Opal, MD Department of Medicine Brown Medical School Providence, RI GUIDELINES FOR LETTERS. Letters discussing a recent JAMA article should be received within 4 weeks of the article’s publication and should not exceed 400 words of text and 5 references. Letters reporting original research should not ex- ceed 500 words and 6 references. All letters should include a word count. Letters must not duplicate other material published or submitted for publication. Letters will be published at the discretion of the editors as space permits and are subject to editing and abridgment. A signed statement for authorship criteria and respon- sibility, financial disclosure, copyright transfer, and acknowledgment is required for publication. Letters not meeting these specifications are generally not consid- ered. Letters will not be returned unless specifically requested. Also see Instruc- tions for Authors (p 104). Letters may be submitted by surface mail: Letters Editor, JAMA, 515 N State St, Chicago, IL 60610; e-mail: JAMA-letters@ama -assn.org; or fax (please also send a hard copy via surface mail): (312) 464-5225. Letters Section Editor: Stephen J. Lurie, MD, PhD, Senior Editor. ©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, January 1, 2003—Vol 289, No. 1 41