Back

Using the shock index to rule out stroke in patients with impaired consciousness in the emergency department: a single-centre retrospective study

Kato, K.; Sagisaka, S.; Ueda, M.; Tamune, H.

2020-05-20 emergency medicine
10.1101/2020.05.16.20101204 medRxiv
Show abstract

BackgroundStroke is a significant cause of impaired consciousness. During a stroke, intracranial pressure increases, resulting in hypertension and bradycardia, also known as the Cushing response. Previous studies showed that systolic blood pressure (sBP) was useful for diagnosing stroke. We hypothesized that stroke patients would show a low Shock Index (SI, pulse rate divided by sBP), and that a high SI would therefore be also useful for ruling out stroke in patients with impaired consciousness. MethodsThis was a single-centre retrospective medical record review conducted between 2015 and 2016. We identified 194 patients with stroke and 588 patients without stroke. We included all out-patients with impaired consciousness who underwent cranial computed tomography (CT) and allocated them to the stroke (cerebral infarction, intracerebral haemorrhage, and subarachnoid haemorrhage) and non-stroke groups. For both the sBP and SI, we calculated the sensitivity and specificity of each cut-off value and compared their diagnostic value. ResultsThe area under the receiver operating characteristic curve was 0.801 (95% Confidence Interval: 0.761-0.840) for the sBP and 0.764 (0.729-0.799) for the SI. For an sBP higher than 90 mmHg and an SI lower than 1, the sensitivity was 99.5% for both. The specificity was 10.9% and 15.1%, respectively, and the negative likelihood ratio was 0.045 and 0.034, respectively. Subgroup analyses suggested that predictive value of SI was enough high regardless of the subtype of stroke, patients past history of hypertension, or level of consciousness. ConclusionSI may be useful for excluding stroke in patients with impaired consciousness. This study suggested that when the pulse rate exceeds sBP, stroke can be ruled out. In these cases, we suggest that physicians focus on non-neurological examinations and treatments rather than diagnostic examinations for stroke.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.