To investigate mortality of ICU patients over a 3-month period after an initial episode of septic shock and to identify factors associated with mortality.
Prospective multicenter observational cohort study.
Fourteen ICUs from 10 French nonacademic and university teaching hospitals.
All consecutive adult patients with septic shock admitted between October 2009 and September 2011 were eligible.
Multivariable analyses were performed using a Cox proportional hazard model and a flexible extension of the Cox model. In total, 1,495 of 10,941 patients (13.7%) had septic shock and 1,488 patients (99.5%) were included. Median age was 68 years (range, 58–78 yr). The majority of admissions (84%) were medical. Median (interquartile range) Simplified Acute Physiological Score II and Sequential Organ Failure Assessment were, respectively, 56 (45–70) and 11 (9–14). ICU and hospital mortality were, respectively, 39.4% and 48.6%. At 3 months, 776 patients (52.2%) had died. Factors significantly associated with increased risk of death in the multivariable Cox model were older age, male sex, comorbidities (immune deficiency, cirrhosis), Knaus C/D score, and high Sequential Organ Failure Assessment score. Flexible analyses indicated that the impact of Sequential Organ Failure Assessment score was greatest early after septic shock, while the onset of the effect of age, nosocomial infection, and cirrhosis was later.
This is the most recent large-scale epidemiological study to investigate medium-term mortality in nonselected patients hospitalized in the ICU for septic shock. Advances in early management have improved survival at the initial phase, but risk of death persists in the medium term. Flexible modeling techniques yield insights into the profile of the risk of death in the first 3 months.
1Service de réanimation médicale, Centre Hospitalier Universitaire, Dijon, France.
2CHU de Dijon, Centre d'Investigation Clinique-Epidémiologie Clinique/Essais Cliniques, Dijon, France.
3Service de réanimation polyvalente, Centre Hospitalier, Haguenau, France.
4Service de réanimation Médicale, Centre Hospitalier Universitaire Hôpital Civil, Strasbourg, France.
5Service de réanimation médicale, Centre Hospitalier Universitaire, Besançon, France.
6Service de réanimation médicale, Centre Hospitalier Universitaire Hautepierre, Strasbourg, France.
7Service de réanimation médicale, Centre Hospitalier Universitaire Hôpital Central, Nancy, France.
8Service de réanimation médicale, Centre Hospitalier Universitaire, Reims, France.
9Service de réanimation polyvalente, Centre Hospitalier, Metz, France.
10Service de réanimation médicale, Centre Hospitalier Universitaire Brabois, Nancy, France.
11Service de réanimation médicale, Centre Hospitalier, Mulhouse, France.
12Service de réanimation médicale, Centre Hospitalier, Vesoul, France.
13Service de réanimation polyvalente, Centre Hospitalier, Belfort, France.
14Service de réanimation polyvalente, Centre Hospitalier, Montbéliard, France.
15Service de réanimation polyvalente, Centre Hospitalier, Colmar, France.
16Institut National de la Santé et de la Recherche Médicale, U866, Dijon, France.
17Service de Biostatistique et d’Informatique Médicale (DIM), Centre Hospitalier Universitaire, Dijon, France.
18Department of Epidemiology and Biostatistics, McGill University, Montreal, Quebec, Canada.
19Université de l’Océan Indien, St. Denis, Ile de la Réunion, France.
A complete list of the EPISS investigators is provided in Appendix 1.
All authors conceived and designed the study. Drs. Pavon, Binquet, and Quenot analyzed and interpreted the study. All authors drafted the manuscript for important intellectual content and approved for submission.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal).
Supported, in part, by the Programme Hospitalier de Recherche Clinique Interregional and the Delegation Interrégionale à la Recherche Clinique of the North-East of France.
Presented, in part, at the 31st International Symposium on Intensive Care and Emergency Medicine, March 24-27, 2011, Brussels, Belgium.
Dr. Abrahamowicz is a James McGill Professor at McGill University. Methodological developments and flexible analyses by Dr. Abrahamowicz were partly supported by the grant # 81275 from the Canadian Institutes for Health Research. The remaining authors have disclosed that they do not have any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org