Severe sepsis in do-not-resuscitate patients: intervention and mortality rates

Severe sepsis is a high-mortality disease, and early resuscitation decreases mortality. Do-not-resuscitate (DNR) status may influence physician decisions beyond cardiopulmonary resuscitation, but this has not been investigated in sepsis.

Among Emergency Department (ED) severe sepsis patients, define the incidence of DNR status, prevalence of central venous catheter placement, and vasopressor administration (invasive measures), and mortality.

Retrospective observational cohort of consecutive severe sepsis patients to single ED in 2009-2010. Charts abstracted for DNR status on presentation, demographics, vitals, Sequential Organ Failure Assessment (SOFA) score, inpatient and 60-day mortality, and discharge disposition. Primary outcomes were mortality, discharge to skilled nursing facility (SNF), and invasive measure compliance. Chi-squared test was used for univariate association of DNR status and outcome variables; multivariate logistic regression analyses for outcome variables controlling for age, gender, SOFA score, and DNR status.

In 376 severe sepsis patients, 50 (13.3%) had DNR status. DNR patients were older (79.2 vs 60.3 years, p < 0.001) and trended toward higher SOFA scores (7 vs. 6, p = 0.07). DNR inpatient and 60-day mortalities were higher (50.5% vs. 19.6%, 95% confidence interval [CI] 15.9-44.9%; 64.0% vs. 24.9%, 95% CI 25.1-53.3%, respectively), and remained higher in multivariate logistic regression analysis (odds ratio [OR] 3.01, 95% CI 1.48-6.17; OR 3.80, 95% CI 1.88-7.69, respectively). The groups had similar rates of discharge to SNF, and in persistently hypotensive patients (n = 326) had similar rates of invasive measures in univariate and multivariate analyses (OR 1.19, 95% CI 0.45-3.15).

In this sample, 13.3% of severe sepsis patients had DNR status, and 50% of DNR patients survived to hospital discharge. DNR patients received invasive measures at a rate similar to patients without DNR status.