Should Adult Emergency Department Patients with Suspicion for Sepsis be Screened for Occult Hypoperfusion at Triage?
Jeffrey P. Green, M.D., Nidhi Garg, M.D., Beverly Kraszewski, D.O., Tony Berger, M.D., Sanjey Gupta, M.D.
Study Objective: Consensus recommendations are to test a serum lactate on adult ED patients with suspicion for infection and ≥ 2 SIRS criteria and enroll patients with occult hypoperfusion (defined as serum lactate >4.0 mmol/L) in Early Goal Directed Therapy (EGDT), a time sensitive intervention. The objective of this study was to determine if testing serum lactate by registered nurses (RNs) at triage would decrease time to indentify occult hypoperfusion when compared to patients screened by Emergency Physicians (EPs) in the main ED.
Methods: Prospective observational cohort. Setting: urban teaching hospital (90,000 annual ED visits) with an active EGDT protocol. Triage RNs at the study site were instructed to test a serum lactate on adult patients with suspected infection and ≥ 2 SIRS criteria. Inclusion criteria: patients ≥ 21 years of age with nurse or provider suspicion for infection and ≥2 SIRS criteria screened for occult hypoperfusion during a 4 month period (8/1-11/30/2007). Exclusion criteria: No serum lactate tested in the ED or no RN or EP suspicion for infection. Statistical Analysis: The primary outcome was time from patient arrival to serum lactate result. Secondary outcomes were proportion of patients with occult hypoperfusion, proportion of patients with ED diagnosis of infection (defined by ICD-9 codes), proportion of patients admitted and proportion admitted to an ICU. Chi-square was used for categorical data and Mann Whitney U for continuous nonparametric data.
Results: 251 adult ED patients with suspected infection and ≥ 2 SIRS criteria were screened for occult hypoperfusion during the study period. Median age 74 (IQR 58-83), 53% female, 43% nursing home residents, mean serum lactate 2.75 mmol/L (95%CI 2.49-3.01). 92% of patients screened were admitted, with 17% admitted to an ICU. 27% of patients were screened by a registered nurse at triage, 73% by EP’s in the main ED. Median time from patient arrival to serum lactate result was 55.8 min (95%CI 49.2 to 66 min) for patients screened at triage vs. 75 min (95%CI 67.2 to 88.8 min) for patients screened by Emergency Physicians (P=0.004). Proportion of patients with occult hypoperfusion was 21.4% (95%CI 12.5-30.4%) for patients screened at triage vs. 21.4% (95%CI 16.0-26.7%) for patients screened by EPs. Patients had an admitting diagnosis of infection in 92.1% (95%CI 86.4-97.8%) of patients screened at triage vs. 81.3% (95%CI 76.4-86.3%) of patients screened by EP’s. The proportion of patients admitted was 89.9% (95%CI 83.5 to 96.3) for patients screened at triage vs. 92.5% (95%CI 89.1 to 95.9%) for patients screened by EP’s. Proportion of patients admitted to the ICU was 14.6% (95%CI 7.1 to 22.1%) for patients screened at triage vs. 17.8% (95%CI 13.0 to 22.7%) for patients screened by EP’s.
Conclusion: In this population patients screened for occult hypoperfusion at triage had a more rapid time to serum lactate result than patients screened by EP’s in the main ED with no significant difference in proportion of patients with occult hypoperfusion, proportion of ED diagnosis of infection, proportion of in-patient and ICU admission.