Impact of endotracheal intubation on septic shock outcome: A post hoc analysis of the SEPSISPAM trial

Abstract

Introduction: The objective of the study to is to determine the characteristics associated with endotracheal intubation in septic shock patients.

Methods: This is a post hoc analysis of the database of the SEPSISPAM study, including patients with septic shock.

Results: Among the 776 patients, 633 (82%) were intubated within 12 hours of study inclusion (early intubation), 113 (15%) were never intubated, and 30 (4%) had delayed intubation. Intensive care units (ICUs) were classified according to frequency of early intubation: early intubation less than 80% of patients (lowest frequency: 7 ICUs, 254 patients), 80% to 90% (middle frequency: 5 ICUs, 170 patients), and greater than 90% (highest frequency: 6 ICUs, 297 patients). Type of ICU, pulmonary infection, lactate greater than 2 mmol/L, lower Pao2/fraction of inspired oxygen ratio, lower Glasgow score, and absence of immunosuppression were independently associated with early intubation. Patients never intubated had a lower initial severity and a low mortality rate. In comparison to patients intubated early, patients with delayed intubation had had fewer days alive without organ support by day 28. Intensive care units with the highest frequency of early intubation had a higher mortality rate in comparison to ICUs with middle frequency of early intubation. A nonsignificant increased mortality was observed in ICU with lowest frequency of early intubation.

Conclusions: Practices regarding the place of endotracheal intubation in septic shock may impact outcome.

Keywords: Endotracheal intubation; outcome; practice variation; retrospective analysis; septic shock.