Daniel Pilsgaard Henriksen MD, PhD Dept. of Clinical Chemistry and Pharmacology, OUH Research Unit of Emergency Medicine, SDU 12 November 2014 Community-acquired sepsis among acutely hospitalized medical patients - Incidence, risk factors, and long-term prognosis

Supervisors  Hanne Madsen, MD Ph.D.  Dept. of Respiratory Medicine, OUH  Court Pedersen, MD Professor DMSci  Dept. of Infectious Diseases, OUH  Annmarie Touborg Lassen, MD Professor Ph.D. DMSci  Dept. of Emergency Medicine, OUH 2

Introduction 3

Definitions 4

Raven MC, Lowe RA, Maselli J, Hsia RY (2013) Comparison of presenting complaint vs discharge diagnosis for identifying “nonemergency” emergency department visits. JAMA 309: 1145–1153. doi: /jama

Problems with discharge diagnoses  Difficult to differentiate between community-acquired and hospital-acquired sepsis  Summary of an entire course of admission  Difficult to differentiate between the severity of sepsis  Tends to underestimate the incidence of sepsis  Tends to identfy the more severely ill severe sepsis patients 6

Aims  Based on symptoms and clinical findings to identify patients admitted to the medical ED at Odense University Hospital in a one- year period (September 2010 – August 2011) we aimed to:  Determine to which degree discharge diagnoses of infection could accurately identify community-acquired infections in an ED setting; and to assess if the sites of infection, baseline patient characteristics and disease severity affect the validity of the discharge diagnoses. (Study I)  Estimate the incidence rates of community-acquired sepsis, severe sepsis, septic shock, and sepsis of any severity. (Study II)  Examine the risk factors for hospitalization with community- acquired sepsis and severe sepsis, and sepsis of any severity in a population-based setting. (Study III)  Examine the association between long-term mortality and community-acquired sepsis, severe sepsis, septic shock, and sepsis of any severity, in a population-based setting. (Study IV) 7

Materiale  Patienter indlagt akut i medicinsk regi OUH 1/ /  Akut Modtageafdelingen  Medicinsk Intensiv afdeling  Døde i skadestuen af formodet medicinsk årsag  Registrering af vitalværdier og andre klinisk relevante data  ankomst  første 24 timer.  30 dage – journalgennemgang  Infektion og fokus

Identifying infections  By structured manual chart review  Based on CDC/NHSN definitions 9

Electronic Patient Records, OUH Laboratory informations systems, OUH Blood Gas Analyzer, Medical ED, OUH Microbiology information system, OUH Danish Civil Registration System Funen Patient Administrative System Danish National Patient Register The Danish National Registry of Alcohol Treatment Odense University Pharmacoepidemiological Database Danish National Cancer Register Cohort of acutely admitted patients Cohort of acutely admitted patients Hospital Based DatabasesPopulation-based Registers SIRS Organ dysfunction Bacteremia 10

11

12

13

Demographic characteristics Study II, III, and IV  Among 8,358 admissions to the medical ED or directly to the medical ICU, 1,713 patients presented with an incident admission of sepsis of any severity within the study period  Median age 72 years (5-95% range: years)  793 (46.3%) were males  728 (42.5%) presented with severe comorbidity  621 (36.3%) with sepsis  1071 (62.5%) with severe sepsis  21 (1.2%) with septic shock. 14

Sepsis of any severity: 731/100,000 pyar (95%CI: ) Sepsis: 265/100,000 pyar (95%CI: Severe sepsis:457/100,000 pyar (95%CI: ) Septic shock: 9/100,000 pyar (95%CI: 6-14) 15 Discharge diagnoses Incidence: /100,000 population Discharge diagnoses Incidence: /100,000 population

Risk factors for Hospitalization with Community-acquired Sepsis – a Population-based Case-Control Study. Henriksen DP, Pottega ̊ rd A, Laursen CB, Jensen TG, Hallas J, Pedersen C, Lassen AT. (In review – Critical Care). 16

Aim  Assess risk factors using symptoms and clinical findings to identify sepsis  Difference in risk factors of sepsis and severe sepsis? 17

Sepsis Adj. OR a [95%CI] Severe sepsis Adj. OR [95%CI] c Male gender 0.89 [ ]1.07 [ ] Age, years 2.15 [ ]3.93 [ ] Age, 85+ years 3.66 [ ]7.84 [ ] Immunosuppression 5.03 [ ]4.45 [ ] Alcoholism-related conditions 2.64 [ ]2.93 [ ] Comorbidities Psychotic disorder 1.35 [ ]2.26 [ ] Neurological 1.90 [ ]1.93 [ ] Respiratory 3.70 [ ]3.29 [ ] Cardiovascular 1.46 [ ]1.65 [ ] Diabetes 1.33 [ ]2.02 [ ] Cancer 1.30 [ ]1.47 [ ] Gastrointestinal 1.39 [ ]1.82 [ ] Renal 0.89 [ ]1.07 [ ] 18

15-64 years65-84 years85+ years Adjusted OR a Male gender0.83 [ ]1.05 [ ]1.55 [ ] Immunosuppression9.54 [ ]4.71 [ ]2.45 [ ] Alcoholism-related conditions3.45 [ ]1.91 [ ]2.54 [ ] Moderate comorborbidity, CCI [ ]3.41 [ ]2.01 [ ] Severe comorbidity, CCI>29.32 [ ]8.12 [ ]3.07 [ ] Table 5 – Adjusted odds ratios a for admission with sepsis of any severity stratified by age in age categories. 19

Conclusions  Several independent risk factors.  A large difference in the risk factors’ strength of association in the different age categories.  No difference in the risk factors’ strength of association when stratifying on sepsis severity. 20

Association between disease severity and long-term mortality in patients hospitalized with sepsis, a population-based cohort study. Henriksen DP, Pottegård A, Laursen CB, Jensen TG, Hallas J, Pedersen C, Lassen AT. (submitted – Critical Care) 21

Aim  Long-term mortality of sepsis of any severity  Difference in long-term mortality of sepsis and severe sepsis 22

Absolute mortality measures Cumulative all-cause mortality, % (95%CI) 30 days Sepsis6.1% ( %) Severe sepsis18.8% ( %) Septic shock38.1% ( %) Sepsis, any severity 14.4% ( %) 3 years Sepsis31.4% ( %) Severe sepsis50.0% ( %) Septic shock 71.4% ( %) Sepsis, any severity 43.5% ( %) 23

Adjusted† HR (95%CI) Intermediate-term mortality days Sepsis3.6 ( ) Severe sepsis7.8 ( ) Sepsis of any severity7.1 ( ) Long-term mortality days Sepsis2.5 ( ) Severe sepsis2.7 ( ) Sepsis of any severity2.8 ( ) days (1 year - 2 years) Sepsis1.7 ( ) Severe sepsis2.2 ( ) Sepsis of any severity2.1 ( ) days (2 years - 3 years) Sepsis2.2 ( ) Severe sepsis2.1 ( ) Sepsis of any severity2.2 ( ) 24

Conclusions  Three years post-sepsis admission  Two-fold higher risk of mortality  Intermediate-term mortality: Sepsis severity matters  Long-term mortality: Sepsis severity does not matter 25

Tak for opmærksomheden