a b s t r a c t

Background and purpose: Following the reorganization of a University Medical Center onto a single campus, an Intensive Care Unit was created within the adult Emergency Department (ED ICU). We assessed the effects of these organizational changes on acute stroke management and the intravenous administration of recombinant tissue plasminogen activator (IV rtPA), as characterized by the thrombolysis rate, Door-to-needle time (DNT) and outcome at 3 months.

Methods: Between October 2013 and September 2015, we performed a retrospective, observational, single- center, comparative study of patients admitted for ischemic stroke and treated with IV rtPA during two 321-day periods (before and after the creation of the ED ICU). All patients with ischemic stroke were included. Multivariable logistic regression models were performed. The DNT was stratified according to a threshold of 60 min. A favorable long-term outcome was defined as a modified Rankin Score <= 2 at 3 months.

Results: A total of 1334 ischemic stroke patients were included. Among them, 101 patients received IV rtPA. The

frequency of IV rtPA administration was 5.8% (39 out of 676) before the creation of the ED ICU, and 9.3% (62 out of 668) afterwards (odds ratio (OR) [95% confidence interval (CI)]: 1.67 [1.08-2.60]; p = 0.02). Additionally, the DNT was shorter (OR [95%CI]: 4.30 [1.17-20.90]; p = 0.04) and there was an improvement in the outcome (OR [95%CI] = 1.30 [1.01-2.10]; p = 0.045).

Conclusion: Our results highlight the benefits of a separate ED ICU within conventional ED for acute stroke management, with a higher thrombolysis rate, reduced intrahospital delays and better safety.

(C) 2017

Introduction

Ischemic stroke (IS) is a leading cause of disability, death and cogni- tive impairment [1]. The management of acute IS remains a major con- cern in routine clinical practice [1]. Thrombolysis via the intravenous administration of recombinant tissue plasminogen activator (IV rtPA)

[2] is an approved treatment for acute IS, and is known to drastically re- duce disability [3]. However, the level of benefit associated with this treatment is highly time-sensitive [4-6].

Given that IV rtPA administration is beneficial within a time window of just a few hours (which includes the time needed for diagnostic workup), only a small subset of IS patients actually receives effective treatment. A European survey found that only 3.3% of IS patients receive IV rtPA, and highlighted large variations from one country to another [7]. In France, no N 2% to 10% of all acute stroke patients currently receive treatment, whereas 25% might be eligible [8,9]. This disparity

* Corresponding author.

E-mail address: [email protected] (L. Puy).

emphasizes the need to shorten the duration of diagnostic and thera- peutic workup. It has been reported that several procedures decrease the door-to-needle time [10,11]. In contrast, the effect of organi- zational and logistic changes at a hospital on the thrombolysis rate has not been studied in detail. Here, we assessed the effects of organization- al and logistic changes on acute stroke management following the creation of an Intensive Care Unit (ICU) inside conventional adult Emergency Department (ED ICU) of a university medical center’s. The study’s primary outcome was the before/after change in the thrombol- ysis rate (i.e. the proportion of IS patients actually receiving IV rtPA). The secondary outcomes concerned changes in the rapidity and safety of care for IS patients receiving IV rtPA.

Materials and methods

Study setting

Amiens University Medical Center (Amiens, France) has a catchment area of 800,000 inhabitants for stroke and 1.9 million for thrombectomy.

http://dx.doi.org/10.1016/j.ajem.2017.01.007

0735-6757/(C) 2017

L. Puy et al. / American Journal of Emergency Medicine 35 (2017) 716–719 717

Prior to 2014, the Center’s medical departments were located on two separate sites – each of which had an ED. Importantly, the main ED was located on the same site as the stroke unit and the neurology and neuroradiology departments. Since 2004, these departments have ap- plied a dedicated protocol for the management of acute stroke patients and thrombolysis. In October 2014, all the medical departments at Amiens University Medical Center were brought together on a single site, following the construction of a new building. This enabled the con- solidation of the two EDs to form a single, 19-room department, togeth- er with the creation of a new 6-room ED ICU within the consolidated ED. This unit was created to manage general critical care patients and nota- bly stroke patients. The ED ICU is operated 24/7 by a dedicated care team, comprising emergency medicine physicians, nurses and nursing assistants with experience of life-threatening emergencies and training in stroke management. All patients admitted for suspected stroke with- in a timeframe compatible with IV rtPA administration were sent to the ED ICU (either directly for patients managed by paramedics or upon ad- mission to the ED by the triage nurse), according to a specific protocol. The objective of this protocol was to accelerate the management of patients and thus increase the thrombolysis rate (i.e. the proportion of eligible patients receiving IV rtPA). To other relevant changes occurred in October 2014. Firstly, a senior duty neurologist was present 24/7; previously, he/she had been present in the ED between 8.00 am and midnight and was on call between midnight and 8.00 am. Secondly, MRI (with direct access from the ED ICU) was promoted as the first- line imaging method. All other aspects of patient management (clinical examinations, laboratory tests, indications and contraindications for thrombolysis) were similar in the two periods.

To assess the effects of these organizational changes, we performed a retrospective, observational, single-center, comparative study of the management of patients admitted for ischemic stroke during two 321- day periods: from October 15th 2013, to September 1st 2014 (i.e. before the creation of the ED ICU), and from October 15th 2014 to September 1st 2015 (i.e. after the creation of the ED ICU).

Population and outcome parameters

During both study periods, we included all patients with a diagnosis of cerebral infarction on discharge from the medical center (I63.0, I63.1, I63.2, I63.3, I63.4, I63.5, I63.6, I63.7, and I63.8).

The following data were extracted from the patients’ medical re- cords: age, gender, initial National Institutes of Health Stroke Scale (NIHSS score), type of brain imaging (non-contrast computed tomogra- phy or MRI), the use of IV rtPA, the time interval between symptom onset and IV rtPA, and the DNT. “Minor stroke” was defined as an NIHSS score <= 4 on admission [12]. Symptomatic intracranial hemor- rhage (ICH) was defined by a visible parenchymal hemorrhage on imag- ing 22-36 h post-treatment, combined with Neurological deterioration (by >= 4 points on the NIHSS) and/or a vigilance disorder [3]. The clinical outcome was assessed in terms of the modified Rankin score at 3 months post-treatment [13]. A favorable outcome was defined as modified Rankin score <= 2 at 3 months.

Statistical analysis

Data were respectively expressed as the mean +- SD, the median [IQR] and the percentage for continuous, ordinal and categorical vari- ables. Continuous variables were compared using a t-test, and ordinal variables were compared using a Mann-Whitney U test. Differences be- tween proportions were tested using Fisher’s test or a chi-squared test. To determine whether the effect of the ED ICU was constant over time, we plotted the cumulative number of IV rtPA administrations be-

fore and after the unit’s creation.

In order to assess the secondary endpoints (the rapidity and safety of care), we performed two logistic regression analyses. The dependent variables were the intrahospital delay (i.e. the DNT <= 60 min versus

N 60 min) and a favorable clinical outcome (modified Rankin score at 3 months <= 2 versus N 2). The independent variables were the creation of the ED ICU, age, gender, initial severity (initial NIHSS score), type of imaging (MRI vs. CT), and treatment delays (the time between symp- tom onset and hospital admission in the first regression analysis, and the time interval between symptom onset and IV rtPA administration in the second one).

All statistical analyses were performed using R software (http:// www.r-project.org/). The threshold for statistical significance was set to p <= 0.05. The study was approved by the local investigational review board.

Results

A total of 96,788 patients attended the ED during the study period as a whole (46,201 before the creation of the ED ICU, and 50,587 after- wards). 1334 patients had experienced ischemic stroke and were in- cluded in the study: 676 (1.5% of total admissions) before the creation of the ED ICU and 668 (1.3%) afterwards (p = 0.06). The thrombolysis rate increased from 5.8% (n = 39) before the creation of the ED ICU to 9.3% (n = 62) afterwards (odds ratio (OR) [95% confidence interval (CI)]: 1.67 [1.08-2.60], p = 0.02). The before/after plots of the cumula- tive number of IV rtPA administrations diverged over time – reflecting the increase in the thrombolysis rate after the creation of the ED ICU (Fig. 1). The two periods did not differ in terms of the demographic and clinical characteristics of the thrombolysed patients (Table 1). Like- wise, the two periods did not differ with regard to the number of strokes of unknown onset and strokes occurring after midnight (n = 4 (10.3%) before the ED ICU’s creation; n = 5 (8.1%) after the ED ICU’s creation; ns). The proportion of minor strokes treated with thrombolysis was higher after the creation of the ED ICU (n = 11, 18.3%) than before (n = 1, 2.6%); p = 0.046. The proportion of DNTs <= 60 min was greater after the creation of the ED ICU (OR [95%CI]: 4.30 [1.17-20.90] in an ad- justed model; p = 0.04). This was still the case after we excluded pa- tients having undergone brain imaging performed in a local hospital prior to referral to our center (before creation of the ED ICU: n = 0; af- terwards: n = 2, ns).

After creation of the ED ICU, MRI was performed more frequently be- fore thrombolysis – thanks to the direct access to the imaging unit from the ED ICU.

Fig. 1. Comparison of the cumulative numbers of IV rtPA administrations before and after the creation of the ED ICU (321 days of follow-up for each period). The dotted line and the solid-line represent the cumulative number of IV rtPA administrations before and after the creation of the ED ICU, respectively. The distance between the two lines continued to grow over time, regardless of the time since the ED ICU’s creation.

718 L. Puy et al. / American Journal of Emergency Medicine 35 (2017) 716–719

Table 1

Demographic and clinical characteristics, the thrombolysis rate, delays and outcomes before and after the creation of the ED ICU.

	Before the creation of the ED ICU (n = 39)	After the creation of the ED ICU (n = 62)	p Value
Demographic and medical characteristics of thrombolysed patients
Age (y), mean +- SD	68.9 +- 16.3	72.5 +- 12.6	0.24
Gender (male), n (%)	19 (48.7)	37 (59.7)	0.38
NIHSS, mean +- SD; median [IQR]	12.4 +- 6.4; 10 [7-18]	10.7 +- 6.9; 8 [6-15]	0.23
Minor stroke, n (%)	1 (2.6)	11 (18.3)	0.046
Stroke of unknown onset, n (%)	1 (2.6)	6 (9.7)	0.33
MRI/CT scan, n (%)	11 (28)/28 (72)	54 (92)/8 (8)	b0.001
Time interval between symptom onset and admission (min), median [IQR]	81.5 [58-103]	89 [70-134]	0.09
DNT? <= 60 min, n (%)	14 (35.9)	28 (46.7)	0.04a
symptomatic intracranial hemorrhage, n (%)	7 (17.9)	3 (4.8)	0.06
Modified Rankin score <= 2 at 3 months, n (%)	17 (44.7)	41 (69.5)	0.045a

Significant results appear in bold.

^a Multivariate logistic regressions, adjusted for age, gender, initial severity, type of neuroimaging and delays.

* Door-to-needle time.

The clinical outcome was better after the creation of the ED ICU inde- pendently of age, gender, time interval between symptom onset and IV rtPA administration and imaging modality (p = 0.045). The proportion of cases with symptomatic ICH was lower after the creation of the ED ICU (corresponding to a 73% reduction in the relative risk).

Discussion

Our present results revealed some important effects of major organi- zational changes in the ED: (a) a higher thrombolysis rate, (b) an in- crease in the proportion of DNTs <= 60 min, and (c) a better clinical outcome.

Following the creation of the ED ICU, we observed an immediate, substantial increase in the number of IV rtPA administrations, which persisted over time. At the end of the follow-up period, the total number of IV rtPA administrations had increased by 60%. The thrombolysis rate rose to 9.3%, which is in the upper range of literature data for French medical centers [8]. The benefit of this organizational change was also emphasized by the improved detection of minor strokes and the reduc- tion in intrahospital delays. The ED ICU is a well-defined operating unit that circumvents the everyday congestion often observed in conven- tional EDs. The 24/7 presence of a specifically trained medical/paramed- ical team in the ED ICU may have substantially increased the efficiency of stroke management. Direct access to an MRI facility may also have contributed to the observed increase in the thrombolysis rate, thanks to better detection of ischemia and the expansion of the indication for IV rtPA to stroke of unknown onset.

After creation of the ED ICU, the median DNT was 66 min; this value is similar to that observed in routine practice in US hospitals [14]. We achieved a DNT b 60 min (the target value in the current French guide- lines) in 47% of cases. In many literature reports, this is achieved in b 30% of patients [15].

The number of cases of symptomatic ICH was smaller (albeit not sig- nificantly) after the creation of the ED ICU. One can speculate that this difference would become statistically significant in a larger sample. Im- portantly, we observed that the rise in the IV rtPA administration rate was associated with a higher proportion of favorable outcomes. The presence of a dedicated medical/paramedical care team doubtless en- abled closer monitoring of blood pressure and glycemia values. The higher proportion of treated minor strokes [12] and the use of MRI may also have contributed to this favorable outcome. Nevertheless, the results of our multivariable logistic regression analysis showed that the ED ICU’s positive effects were independent of the initial Stroke severity and the type of imaging.

The present study’s limitations included the relative small sample size, and the retrospective, observational, single-center design (resulting in moderate external validity). One of the greatest challenges in stroke management is the achievement of homogeneous intrahospital delays [16]. Thus, we sought to perform medical and

paramedical expertise in stroke management. To this end, we intend to organize a simulated “code stroke” session for all the stakeholders in acute stroke management [17-19]. In view of a lack of data, we could not investigate differences in pre-hospital management between the two periods.

Our study also had some strengths. Our data are representative of a typical patient population in an acute stroke unit, with baseline charac- teristics similar to those reported in the literature [3]. To the best of our knowledge, the present study is the first to have evaluated the impact of organizational change in a consolidated university medical center ED on acute stroke management.

In conclusion, the creation of an ED ICU within a university medical center ED gave us a unique opportunity to evaluate the impact of such organizational changes on acute stroke management. Our results are ex- tremely promising and argue in favor of modernizing the management of neurovascular diseases. Although considerable progress has been made, substantial challenges remain; efforts must be made to further improve the quality of care and, in particular, reduce the time to treatment.

Funding sources/disclosures

None.

Acknowledgments

None.

References

1. Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE, et al. Population- based study of event-rate, incidence, case fatality, and mortality for all acute vascular events in all arterial territories (Oxford Vascular Study). Lancet Lond Engl Nov 19 2005;366(9499):1773-83.
2. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med Dec 14 1995;333(24):1581-7.
3. Wahlgren N, Ahmed N, Davalos A, Ford GA, Grond M, Hacke W, et al. Thrombolysis with alteplase for acute ischaemic stroke in the safe implementation of thrombolysis in stroke-monitoring study (SITS-MOST): an observational study. Lancet Lond Engl Jan 27 2007;369(9558):275-82.
4. Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Broderick JP, et al. Associa- tion of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet Lond Engl Mar 6 2004;363(9411):768-74.
5. Lees KR, Bluhmki E, von Kummer R, Brott TG, Toni D, Grotta JC, et al. Time to treat- ment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet Lond Engl May 15 2010; 375(9727):1695-703.
6. Saver JL, Fonarow GC, Smith EE, Reeves MJ, Grau-Sepulveda MV, Pan W, et al. Time to treatment with intravenous tissue plasminogen activator and outcome from acute ischemic stroke. JAMA Jun 19 2013;309(23):2480-8.
7. Leys D, Ringelstein EB, Kaste M, Hacke W. Facilities available in European hospitals treating stroke patients. Stroke J Cereb Circ Nov 2007;38(11):2985-91.

   L. Puy et al. / American Journal of Emergency Medicine 35 (2017) 716–719 719

   Bejot Y, Rouaud O, Benatru I, Fromont A, Couvreur G, Caillier M, et al. Contribution of the Dijon Stroke registry after 20 years of data collection. Rev Neurol (Paris) Feb 2008;164(2):138-47.

8. Yip TR, Demaerschalk BM. Estimated cost savings of increased use of intravenous tis- sue plasminogen activator for acute ischemic stroke in Canada. Stroke J Cereb Circ Jun 2007;38(6):1952-5.
9. Strbian D, Ahmed N, Wahlgren N, Lees KR, Toni D, Roffe C, et al. Trends in door-to- thrombolysis time in the safe implementation of stroke thrombolysis registry: effect of center volume and duration of registry membership. Stroke J Cereb Circ May 2015;46(5):1275-80.
10. Meretoja A, Weir L, Ugalde M, Yassi N, Yan B, Hand P, et al. Helsinki model cut stroke thrombolysis delays to 25 minutes in Melbourne in only 4 months. Neurology Sep 17 2013;81(12):1071-6.
11. Strambo D, Zambon AA, Roveri L, Giacalone G, Di Maggio G, Peruzzotti-Jametti L, et al. Defining minor symptoms in acute ischemic stroke. Cerebrovasc Dis Basel Switz 2015;39(3-4):209-15.
12. Sulter G, Steen C, De Keyser J. Use of the Barthel index and Modified Rankin scale in acute stroke trials. Stroke J Cereb Circ Aug 1999;30(8):1538-41.
13. Fonarow GC, Zhao X, Smith EE, Saver JL, Reeves MJ, Bhatt DL, et al. Door-to-needle times for tissue plasminogen activator administration and clinical outcomes in

    acute ischemic stroke before and after a quality improvement initiative. JAMA Apr 23 2014;311(16):1632-40.

    Fonarow GC, Smith EE, Saver JL, Reeves MJ, Bhatt DL, Grau-Sepulveda MV, et al. Timeliness of tissue-type plasminogen activator therapy in acute ischemic stroke: patient characteristics, hospital factors, and outcomes associated with door-to-nee- dle times within 60 minutes. Circulation Feb 22 2011;123(7):750-8.

14. Sauser K, Levine DA, Nickles AV, Reeves MJ. Hospital variation in thrombolysis times among patients with acute ischemic stroke: the contributions of door-to-imaging time and imaging-to-needle time. JAMA Neurol Sep 2014;71(9):1155-61.
15. Sung S-F, Tseng M-C. Code stroke: a mismatch between number of activation and number of thrombolysis. J Formos Med Assoc Jul 2014;113(7):442-6 (Taiwan Yi Zhi).
16. Garcia-Santibanez R, Liang J, Walker A, Matos-Diaz I, Kahkeshani K, Boniece I. Com- parison of stroke codes in the emergency room and inpatient setting. J Stroke Cerebrovasc Dis Aug 2015;24(8):1948-50.
17. Chen C-H, Tang S-C, Tsai L-K, Hsieh M-J, Yeh S-J, Huang K-Y, et al. Stroke code im- proves intravenous thrombolysis administration in acute ischemic stroke. PLoS One 2014;9(8):e104862.