department of a university hospital in Hong Kong. MLC Yang. , CA Graham ... Emergency Medicine and Intensive Care Medicine, 30 Gascoigne. Road, Kowloon ...
Hong Kong Journal of Emergency Medicine
Outcome after implementation of sepsis guideline in the emergency department of a university hospital in Hong Kong
MLC Yang
, CA Graham
, TH Rainer
Introduction: Internationally, standard care of patients with severe sepsis consists of early detection, early antimicrobial therapy, and aggressive intravenous fluid therapy to maintain tissue oxygenation and perfusion. In this retrospective study, we aimed to examine the management of patients with severe sepsis in a local university hospital emergency department before and after the implementation of a sepsis management guideline. Method: We collected data on the management and outcome of patients during a three-month period before the implementation of a sepsis guideline (October-December 2009). We then collected similar data one year after the implementation (October-December 2010). Key sepsis management areas and inhospital mortality rates were compared, as were length of resuscitation, three-month mortality rate, hospital length of stay (LOS) and intensive care unit (ICU) LOS. Results: Data from 115 patients were collected in the pre-implementation group, while data on 102 patients were collected for the post-implementation group. There were more patients with hypoperfusion in the post-implementation cohort (25.2% vs. 40.2%, p=0.019). There was no difference in background characteristics, average lactate value, average MAP or number of hypotensive patients between the two groups. Significantly more antibiotics were given after the intervention (13.0% vs. 23.5%, p=0.045) and more patients had a lactate level measured (43.0% vs. 73.5%; p60 yo, alcoholic, altered immunity, DM, active cancer Must add vancomycin (in lieu of Unasyn) for infected shunts, recent head injury or craniofacial surgery. Immediate surgical debridement Use Tazocin 4.5 g in lieu of Cefriaxone if recent hospitalisation, recent antibiotic use, surgical wound infection. Tazocin 4.5 g if IVDU
Blood and urine for culture, bile for culture if available
Blood and urine for culture
Blood, urine and sputum cultures Separate blood culture from each hickmen or central lines Blood and CSF culture
Blood, pus and wound aspirate for culture
Blood and joint aspirate for culture
168
Hong Kong j. emerg. med. Vol. 22(3) May 2015
demographic characteristics and presenting vital signs. Significantly more patients in the post implementation group had hypoperfusion (29 vs. 41, 25% vs. 40%; p=0.019), while there was no difference in the lactate values (mean [±SD] 2.85±2.4 vs. 3.1±2.6 mmol/L; p=0.450), presenting MAP (93±25 vs. 92±26 mmHg; p=0.770) and patients with MAP less than 65 mmHg (17 vs. 17; 15% vs. 17%; p=0.700).
There was no difference in overall patient outcome measurements (mortality rate, LOS). A survival benefit for the subgroup of patients with hypoperfusion was noted (14/29 vs. 11/41; 48.2% vs 27.0%; p=0.060), but this did not reach statistical significance.
Primary and secondary outcomes
In a post-hoc subgroup analysis of patients who presented with hypoxaemia (i.e. SpO2
