Hospitalist-Based Management of Sepsis in an Intermediate Care Unit. Results and Analysis of Factors Related to Poor Prognosis and Need for Discharge to an Intensive Care Unit

Background:

The Emergency Room departments and the Intensive Care Units (ICU) remain the main locations for the treatment of patients with severe sepsis and septic shock. Nevertheless, health-care systems may have limited resources and insufficient ICU beds for attending patients who might benefit from more intensive monitoring than in a general ward. In this context, Intermediate Care Units (ImCU) could reduce costs and improve ICU utilization for sicker patients. The purposes of this study are: to analyse the outcome of patients with sepsis attended in an ImCU led by hospitalists, and to describe factors related to in-hospital mortality and need for ICU discharge.

Methods:

A retrospective observational study, with data collected from April 2006 to December 2013 in a single academic medical centre in Pamplona, Spain, was made. Several variables obtained at admission and during the first 24 hours of ImCU stay were recorded. SAPS 2, SAPS 3 and Sepsis Severity Score (SSS) were also calculated.

Results:

A total of 150 consecutive patients were included. Overall in-hospital mortality was 26% (39/150) with expected mortality rates calculated by SAPS II, SAPS 3 and SSS of 29.2%, 54.9% and 38.2%, respectively. Main factors related to in-hospital mortality were do-not-resuscitate orders, solid organ metastatic cancer, previous history of cirrhosis, functional status measured by ECOG scale, previous length of stay in hospital, increased heart rate, need of non-invasive mechanical ventilation, low pCO2, increased urea, increased bilirubin and prolonged prothrombin time at ImCU admission. SAPS II showed adequate calibration (χ²=8.95, p=0.347) and better discrimination than SAPS 3 and SSS for predicting mortality, with an AUROC of 0.768.

Regarding transfer to a higher level of care, 17% of patients (17/102 after excluding those with do-not-resuscitate orders at ImCU admission) were discharged to the ICU. Main factors related to discharge to the ICU were those related with severity of sepsis such as renal failure, low pH values, prolonged prothrombin time and impaired liver function. SSS showed excellent calibration (χ²=8.95, p=0.347) with better discrimination (AUROC of 0.867) than SAPS II and SAPS 3 to determine patients at risk of need of additional support in the ICU. A cut-off value of SSS of 70 points showed a Sensitivity of 76.5% and a Specificity of 80% for predicting ICU transfer in our population. 

Conclusions:

ImCUs arise as an attractive alternative for the management of selected septic patients, with mortality ratios comparable with conventional intensive care.

Cost-effective studies of intermediate care, and more reliable and accurate scores for ImCU patients with sepsis based on larger and prospective trials, are needed. Future studies should also try to clearly define and protocolize admission criteria of patients with sepsis to different areas like general ward, ImCUs and ICUs.