The study was designed to assess whether high dosages of norepinephrine are associated with increased death rate and to determine the dosage of norepinephrine associated with an intensive care unit (ICU) death rate greater than 90%. We conducted a retrospective, noninterventional, observational study in a single ICU (15 beds) of an academic hospital. From January 2009 to May 2013, data of all patients with a diagnosis of septic shock were extracted from our database. Data were collected at the time of the admission in ICU, at the onset of septic shock, and when the maximal posology of norepinephrine was reached. Mortality was assessed in ICU, in hospital, and at day 90. Among the 324 patients with septic shock, the death rate was 48%. The death rate reached 90% for the quantile of patients receiving more than 1 μg/kg per minute of norepinephrine. In our cohort, four independent factors associated with mortality were identified: age (odds ratio, 1.02 [95% confidence interval, 1.00-1.04]; P = 0.02), thrombocytopenia (odds ratio, 3.8 [95% confidence interval, 1.8-8.5]; P < 0.001), urine output less than 500 mL (odds ratio, 8.7 [95% confidence interval, 3.6-25]; P < 0.001), and dosage of norepinephrine greater than 1 μg/kg per minute (odds ratio, 9.7 [95% confidence interval, 4.5-23]; P < 0.001). However, because of the study's design, unmeasured confounding factors should be taken into account in our findings.

In our cohort, the initial imaging assessment allowed appropriate decisions in 755 of 756 patients, with a global therapeutic impact of 10%. The pelvic x-ray had a minimal therapeutic impact, and in the patients with normal lung ultrasounds, the chest x-ray marginally affected the management of our patients. The potential consequences of abandoning systematic chest and pelvic x-rays should be investigated in future randomized prospective studies.