Adding eosinophil count to EMERALD rules predicts subarachnoid haemorrhage better in emergency department.

Background: Subarachnoid haemorrhage (SAH) clinical decision rules can provide successful results in the differential diagnosis of non-traumatic headache.

Aim: The aim of this study is to investigate whether a laboratory parameter that can be added to clinical decision-making rules can better predict subarachnoid haemorrhages in patients presenting to the emergency department with headache.

Methods: We conducted a retrospective cohort study between March 2017 and March 2019. Patients over the age of 18 who admitted to the emergency department with non-traumatic, acute headache last 14 days before admission and evaluated by imaging and laboratory studies were included in the study.

Results: A total of 867 patients were included and 141 of them had SAH. In detecting the SAH, Ottawa SAH rule sensitivity (85.1%), specificity (47.1%), positive predictive value (PPV) (23.8%) and negative predictive value (NPV) (94.2%), and for EMERALD SAH, rule sensitivity (96.4%), specificity (43.2%), PPV (24.8%) and NPV (98.4%). In the regression analysis, statistically significant result was obtained to exclude the diagnosis of SAH with the decrease in the eosinophil value (cutoff < 0.085 10^3/µL, OR: 0.011, 95% CI: 0.001-0.213, p = 0.003). When eosinophil value was added to EMERALD SAH rule, it provided a 100% of sensitivity, a 38.4% of specificity, a 24% of PPV and a 100% of NPV in detecting the SAH.

Conclusions: EMERALD SAH rule plus eosinophil, which offers 100% sensitivity and NPV for predicting SAH in adult non-traumatic headaches, may be recommended as a successful and practical decision rule for clinical use according to the Ottawa and EMERALD SAH rule.