Send letter to journal:
Re: Usefulness of MEDS score for risk stratification of severe sepsis medical patients

Dear Sir,

We read with interest the article by Chen et al which demonstrated that a MEDS score greater than 12 may be discriminant for Emergency Department (ED) triage, ICU placement (1). These results of importance are in accordance with those of our group. We prospectively evaluated the accuracy of the MEDS score in 286 patients with documented sepsis admitted into our medical ICU (10 beds) after an ED consultation and we compared it with the performance of the SAPS II score calculated with the worst values of physiologic variables during the first 24 hr of ICU admission. The median age (interquartile) was 69 yr (52.1-78.8), for a median MEDS value of 10 (8-12) and a SAPS II score of 51 (35.7-70). Community-acquired pneumonia was the main source of sepsis (58.6%). In our series, the two scores discriminate well between survivors and non- survivors (in-hospital mortality rate: 32.9%) if we consider the area under the ROC curve which was 0.771 [95% confidence interval (CI), 0.714-0.828) for the MEDS score and 0.830 (95%CI, 0.781-0.880) for the SAPS II score. Furthermore, severity of illness, age, and mortality increased when stratifying patients into the five risk groups defined by Shapiro et al (p<. 000.1) (see table) (2). Whereas, Chen et al reported an increased mortality in case of MEDS score > 12 [calculated odds ratio (OR): 4.51], in our population a cut-off value of 10 was already an important predictor of death: OR, 5.4(95%CI: 3.01-9.72), sensitivity 0.475, specificity 0.857, positive predictive value 0.809, negative predictive value 0.562. As in Chen et al cohort, the MEDS score performs well and then may be an appropriate tool in identification of ED patients at high risks of death from sepsis. However, its usefulness should be tested in surgical ICU patients and prospectively evaluated in the ED to decide ICU admission.

Table 1

References

1 Chen CC, Chong CF, Liu YL, et al. Risk stratification of severe sepsis patients in the emergency department. Emerg Med J 2006; 23: 281-5