Objectives & Background A previous audit has shown that in general with each rise in NEWS group there is an increase in serum lactate measured. However, it was noted that in the lowest NEWS Group 0–4 the average lactate was 2.5. A lactate level of >2 has been used as an initial marker of severity of illness yet these patients would be missed at triage.
We decided to look at whether adding the lactate to the NEWS group would improve the early identification of septic patients at triage.
A recent paper has used this in the context of patients with pneumonia.
Methods We retrospectively reviewed 200 cases of adult patients who presented to the ED who had been identified as septic and had a sepsis 6 form completed. Initial NEWS scores and serum lactate levels were noted.
Results 200 cases were identified over a 6 month period–109 male; 91 female; age range 18–95 yrs, average 66.7 yrs. 3 notes were excluded as a serum lactate was not taken.
4 patients were admitted to ICU; 17 admitted to HDU and 22 patients died.
NEWS scores ranged from 1–19, average 7.4; IQR 1–NEWS 5; IQR 3–NEWS 9.
Initial serum lactate levels ranged from 0.6–15.4, average 2.7; IQR1–1.4; IQR 3– 3.
NEWSL scores ranged from 2.5–27.4, average 10.2; IQR1 –7.4; IQR 3– 12.3.
The paper by S Jo et al combined lactate with NEWS scores and used ranges –<3; 3.1–5.2; 5.3–8; and 8.1+. We have adopted the same ranges.
Numbers of patients are shown in table 1.
Table 2 shows the number of patients in each NEWS group who required critical care input or died. With each rise in NEWS group, critical care patients are identified. However, 1 patient died in the NEWS 0–4 group which would not be deemed high risk at triage.
Table 3 shows the number of patients in each NEWSL group who required critical care input or died. When adding the lactate to the NEWS score all patients were identified that required critical care or died in the highest NEWSL 8.1+ group. This group takes into account the patient in the NEWS 0–4 group who died.
The ROC curve for NEWS vs NEWSL is shown in figure1. The AUC is 0.7187 with an accuracy of 0.6617 which is deemed fair discriminatory ability for the NEWSL score.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.