Elsevier

Clinical Imaging

Volume 45, September–October 2017, Pages 65-70
Clinical Imaging

Original Article
The impact of installing an MR scanner in the emergency department for patients presenting with acute stroke-like symptoms

https://doi.org/10.1016/j.clinimag.2017.05.015Get rights and content

Highlights

  • After installation of the MR in the ED, there was increased use of the ED MR.

  • More patients admitted actually had a stroke, so LOS did not decrease.

  • Patients discharged after a negative MR were not at risk for stroke within 60 days.

  • ED MR yields improved efficiency and accuracy without under-diagnosing strokes.

Abstract

Purpose

We examined the impact of an MR scanner in the emergency department (ED) on ED length of stay (LOS), hospital (H) LOS, hospital admission rate, hospitalization costs, and ED re-presentation rate of patients presenting with stroke-like symptoms (SLS). We hypothesized that the ED MR would increase efficiency of patient care.

Methods

The number of MRIs performed in the ED vs. inpatient setting, EDLOS, HLOS, hospitalization charges, admission rate, discharge diagnoses, and 30–60-day ED re-presentation rates were determined for ED patients with SLS six months before (2011) and after (2012) ED MR installation.

Results

362 and 448 patients with SLS presented to the ED, and 196 and 176 patients were admitted in 2011 and 2012 respectively. In 2011, 36 (18.4%) admitted patients, and, in 2012, 68 (38.6%) had MRIs in the ED, p < 0.001. In 2011, 74 (37.8%) admitted patients were diagnosed with ischemic stroke, compared to 92 (52.3%) in 2012, p = 0.007. HLOS was longer and charges higher for patients with stroke. No patients returned with a confirmed diagnosis of CVA or TIA within 0–60 days after being discharged from the ED with negative MR.

Conclusions

With the ED MR, more admitted patients 1) got scanned in the ED and 2) were diagnosed with stroke. Because this led to more patients on the stroke service actually suffering from strokes (and not other diagnoses), the overall HLOS and charges of patients presenting with SLS were not reduced by ED MR screening. Discharge after a negative ED MR did not incur risk of TIAs or strokes over the ensuing 60 days. Therefore, not only does a dedicated MR scanner in the ED aid in the acute diagnosis of a CVA or other neurologic disorder, but it does so without the risk of under-diagnosing TIAs or evolving strokes in the presence of a negative MRI.

Introduction

MRI is widely established as invaluable in the assessment of patients with stroke-like symptoms (SLS), and recently has been increasingly used in the emergency department (ED) setting [1], [2]. The major diagnostic question for patients who present with new neurological signs or symptoms is whether these are related to acute cerebrovascular disease - ischemic or hemorrhagic - or whether the symptoms are due to non-cerebrovascular causes. MR diffusion-weighted imaging (DWI) is capable of demonstrating ischemic brain tissue changes within minutes of the insult, well in advance of detection by non-contrast CT [3], [4], [5], [6]. Prior studies have demonstrated MRI's superiority in the initial evaluation of patients with SLS and have called for utilization of MRI over CT in the acute setting [5], [6], [7], [8], [9], [10].

At our institution, an MR scanner was installed in the ED in early 2012 and became operational in April 2012. Prior to 2012, ED patients who needed MRI were assessed on inpatient scanners; however, this was performed on a limited basis prior to patient admission to the hospital. Since the installation of the scanner, we have observed an increased number of ED MRI exams ordered by ED physicians or stroke neurologists as a part of stroke work-up, at times preceding or completely replacing the non-contrast CT.

Our early experience with the in-ED scanner suggested that more patients were being “ruled out” for stroke through the rapid and early evaluation by the ED MR scanner and were thereby being discharged from the ED with a “non-stroke” diagnosis rather than being worked up as an inpatient. The stroke literature states that patients are at an increased risk of ischemic stroke in the time period immediately following TIA, with that risk being equal to 3.5% at 2 days, 5.2% at 7 days, 8.0% at 30 days, and 9.2% at 90 days after TIA [11], [12]. Given these data, a concern was raised that this early evaluation in the ED might lead to patients with cerebral ischemic symptoms being discharged from the ED while still at risk for subsequent stroke after a negative ED MRI.

We hypothesized that with the installation of the ED MR scanner, more admitted patients would get MRIs in the ED, and therefore the EDLOS would be longer (due to the added time of MRI examination), and HLOS would be shorter (due to a more timely “in-ED” patient evaluation and resulting appropriate early management). We also believed that the percentage of patients diagnosed with cerebrovascular ischemia admitted to the stroke service would be higher (because ED MRIs would screen out non-vascular pathologies before admission). We hoped that the rates of re-presentation to the ED and admission following a negative stroke work-up in the ED would be negligible.

Section snippets

Methods

In compliance with the Health Insurance Portability and Accountability Act (HIPAA), this study protocol was reviewed and approved by the Internal Review Board (IRB). Informed consent was waived due to the retrospective nature of the study.

Summary statistics

During the time interval from 7/1/2011–12/31/2011 (prior to scanner installation), 31,032 patients were evaluated in the ED and 362 patients (mean age: 52.9 years; age range: 19–94; male: 156) presented with SLS (2011 cohort). During the time interval from 7/1/2012–12/31/2012 (after scanner installation), 33,299 patients were evaluated in the ED and 448 patients (mean age: 50.9 years; age range: 7–95; male: 174) presented with SLS (2012 cohort). In the 2011 cohort, 196 (54.1%) of these patients

Discussion

Since installation of the MR scanner in the ED at our institution, a greater number of patients who are admitted with SLS underwent MRI examination in the ED setting (38.6% vs. 18.4%, p < 0.001). However, the overall proportion of the patients who got MR imaging as a part of stroke work-up did not change and remained at approximately 81%. Compared to 2011, in 2012, we observed a significantly higher number of patients admitted with SLS who were ultimately diagnosed with ischemic stroke (52.3% vs.

Conclusions

After installation of the MR scanner in the ED, there was increased utilization of the ED MR for patients presenting with SLS. ED MRs replaced those performed as an inpatient, so overall MRs/patient did not increase. More patients admitted on the stroke service actually had a stroke, therefore hospital length of stay for all patients did not decrease since patients without strokes are discharged more rapidly than those admitted with strokes. Patients discharged from the ED after a negative ED

Acknowledgments

None

References (21)

There are more references available in the full text version of this article.

Cited by (0)

View full text