Introduction and Aim Sepsis and pneumonia commonly present together, however the preferred clinical scoring system to use is unclear. NEWS (National Early Warning Score) and SIRS (Systemic Inflammatory Response Syndrome) and SOFA(Sequential Organ Failure Assessment) scores can be used to evaluate patients admitted with sepsis. Due to complexity of SOFA scoring, qSOFA has been proposed for routine practice.¹ We compared sepsis scores in patients with community acquired pneumonia related sepsis (CAP sepsis) and compared this to sepsis from other causes (non-CAP sepsis). We also evaluated how sepsis scoring systems compared with the CURB-65 score to predict mortality and ICU admission in CAP sepsis.

Methods Medical records were audited between 01/09/2016–20/03/2017 at Barking Havering and Redbridge University Hospitals NHS Trust. Adult patients were included if ICD-10 codes on discharge/death were Sepsis A40/A41. Physiological and blood parameters were collected at time of trigger from medical/electronic records. All plain chest x-rays (CXR) and CT-chest scans were reviewed independently by two respiratory registrars to confirm consolidation. We calculated the positive predictive value (PPV) and negative predictive value(NPV) of each sepsis scoring system against two outcome measures; i)ICU Admission <24 hours ii)mortality(death) at 30 days.

Outcomes/Results We identified 114 cases of sepsis. Median age-78 years. Male:female ratio 56:58. We found 22/114 (19%) of patients had a diagnosis of CAP sepsis with CXR/CT-chest confirming consolidation <48 hours from admission. Scores for CAP sepsis (n=22) triggered more often than non-CAP sepsis(n=92) for NEWS ≥4 (77% vs 54%,p=0.05), SOFA ≥2 (86% vs 54%,p<0.01). For CAP sepsis and non-CAP sepsis, the PPV and NPV for each sepsis score including CURB-65 were calculated for <24 hour ICU admission and 30 day mortality (Table 1). In patients with CAP sepsis, scores for NEWS/SIRS/SOFA had a high PPV for ICU admission and mortality. CURB-65 ≥3 had a low PPV but higher NPV.

Conclusion Patients with CAP sepsis triggered more often with NEWS ≥4 and SOFA≥2 compared to non-CAP sepsis. Our data suggests that NEWS ≥4,SIRS≥2,SOFA≥2 or CURB-65≥2 may be used as initial screening for CAP sepsis. However, all had low NPV for ICU admission and 30 day mortality. Further studies are needed to evaluate the best scoring system to assess clinical severity for CAP sepsis and avoid unnecessary duplication.

Reference

1. Singer Met al. Third International Consensus Definitions for Sepsis and Septic Shock: JAMA2016;315(8):801–810.