a b s t r a c t

Background: Door-to-needle time (DNT) is a critical consideration in emergency management of acute ischemic stroke . Deficiencies in the widely applied standard hospital workflow process, based on international guidelines, impede Rapid treatment of AIS patients. We developed an in-hospital stroke system to reduce DNT and optimize hospitals’ emergency procedures.

Objectives: To investigate the effect of the in-hospital stroke system on the hospital workflow for AIS patients. Methods: We performed a retrospective study on AIS patients between June 2017 and December 2021. AIS cases were assigned to a pre-intervention group (before the in-hospital stroke system was established) and a post-intervention group (after the system’s establishment). We compared the two groups’ demographic fea- tures, clinical characteristics, treatments and outcomes, and time metrics data.

Results: We analyzed 1031 cases, comprising 474 and 557 cases in the pre-intervention and post-intervention groups, respectively. Baseline data were similar for both groups. Significantly more patients in the post- intervention group (41.11%) were treated with intravenous thrombolysis (IVT) or Endovascular therapy (ET) compared with those in the pre-intervention group (8.65%) (p < 0.001). DNT was markedly improved (decreas- ing from 118 (80.5-137) min to 26 (21-38) min among patients in the post-intervention group treated with IVT or bridging ET. Consequently, a much higher proportion of these patients (92.64%) received IVT within 60 min compared with those in the pre-intervention group (17.39%) (p < 0.001). Consequently, their hospital stays were shorter (8 [6-11] days vs. 10 [8-12] days for the pre-intervention group; p < 0.001), and they showed improved National Institutes of Health Stroke Scale scores at discharge (-2 [-5-0] vs. -1 [-2-0], p < 0.001).

Conclusion: DNT was significantly reduced following implementation of the in-hospital stroke system, which contributed to improved patient outcomes measured by the length of hospital stay and NIHSS scores.

(C) 2023

1. Introduction

In recent years, cerebrovascular diseases, which have a continually rising incidence rate, have emerged as the third leading cause of death [1]. Patients lose their ability to work and require long-term home care and rehabilitation, which induces a heavy Economic burden on their families and society [2]. Every minute saved in the timeline of

* Correspondence to: Yimin Zhu, Poisoning Research Laboratory, Hunan Provincial People’s Hospital, 61 Jiefangxi Road, Changsha 410005, China.

?? Correspondence to: Jun Liu, Changsha Central Hospital, 161 Shaoshannan Road, Changsha 410018, PR China.

E-mail addresses: [email protected] (Y. Zhu), [email protected] (J. Liu).

symptom onset-to-treatment time (OTT) and receipt of Intravenous thrombolysis or endovascular therapy (ET) could result in an average gain of 4.2 extra days of healthy living. Moreover, every 20 min reduction in treatment delay could yield the equivalent of three months, on average, of disability-free life [3]. Therefore, it is imperative to restore the cerebral blood flow as soon as possible. According to the 2018 AHA/ASA guide, acute ischemic stroke patients should receive IVT within 4.5 h after symptom onset [4]. A primary goal of achieving door-to-needle time within 60 min for >50% of AIS patients treated with IVT alteplase should be established [5]. However, despite the existence of international guidelines on stroke care, the emergency procedure requires optimization to achieve a rapid hospital workflow for AIS patients. A stroke information system would

https://doi.org/10.1016/j.ajem.2023.04.008

0735-6757/(C) 2023

be an alternative method to optimize hospital workflow for AIS. In a comparatively earlier stage of the stroke information system, the brain salvage was activated immediately after screening at triage via a com- puterized physician Order entry system [6], through which the orders are immediately delivered to all staff members. The STRAUMA activa- tion [7] is a bundled package that is initiated on patient arrival and al- lows prompt and concurrent evaluation by team members, who are present on time or immediately thereafter and collaborated to make treatment plan. Recently stroke information system employs several functionalities, including group alert to inform all team members of ar- riving patient; image viewer to display medical images on smartphone and tablet platforms [8]. However, a deficiency of the traditional stroke information system is a lack of strict time record method, which limits to discover delayed time metrics in the emergency procedure.

Thanks to the advancement of information technology (IT), it has made easier to know the locations of patients, continuously monitor emergency procedures, and share patient data in real time. In August 2019, an in-hospital, IT-based stroke system combined with a delayed time identification technique was established at Hunan Provincail People’s Hospital(Changsha city, Hunan province, China) to optimize the emergency procedure applied to AIS patients. We conducted a Retro- spective analysis using the medical records of inpatients treated before and after the implementation of the in-hospital stroke system. Our aim was to confirm the effectiveness of the in-hospital stroke system in optimizing the hospital workflow for AIS patients.

1. Methods
   1. Selection of patients

A retrospective study was performed from June 2017 to December 2021 in Hunan Provincial People’s Hospital (The First Affiliated Hospital of Hunan Normal University), which is a 2000-bed facility catering to

>20,000 inpatients annually. According to the AHA/ASA guidelines (2018 edition), we selected patients who met the AIS diagnostic criteria as follows: (a) acute onset; (b) with symptoms or signs of Neurological deficits; (c) confirmation of AIS on brain CT/MRI; (d) exclude cerebral hemorrhage and non-vascular causes. The exclusion criteria were as fol- lows: (a) stroke happened in the hospital; (b) patients who received IVT or ET at other hospitals; (c) patients who had undergone evaluatory brain imaging prior to admission, and (d) missing data.

• 1. Methodology

The in-hospital stroke system, which was established in August 2019, comprises two core components. The first is a technology for sharing information in real time.

Through a mobile phone-based terminal application of the in- hospital stroke system, the patient’s medical information (demographic features, clinical characteristics, brain Imaging results, and Laboratory test results) can be synchronously and simultaneously shared with all stroke team members, who receive pre-notifications. The second com- ponent comprises radiofrequency identification (RFID) and the collec- tion of time metrics. RFID is a method of identifying unique items using radio waves. A basic RFID system comprises a tag (e.g., a wrist- watch) and a reader (with an antenna) (Fig. 1a and b). The reader sends and receives the radio waves. The tag comprises a mounted data-storing chip and antenna. The RFID in our hospital was installed at the gate of the emergency room, the CT room, and the angiography suite. The time metrics of the emergency procedure are automatically collected and recorded in the system when a patient wearing a RFID wristwatch passes by. Therefore, any delay in the emergency procedure would be promptly identified. A software, for error correction, was combined with RFID, which could fill up automatically the missed infor- mation of the tag while identifying. Moreover, if identification completely had failed, which seldomly happen, only need to do were to repeat the identification The two core components combined each other, mean while, opening port and exchanging data with the electronic medical record system, therefore, team members know the timeline of workflow and sharing information in real time.

The selected AIS cases were assigned to one of two groups: a pre- intervention group of patients who received treatment before the in- hospital stroke system’s establishment and a post-intervention group of patients who received treatment after its establishment. Patients in the pre-intervention group were generally managed according to a structured stroke care pathway as follows:

• • 1. Patient arrival. Emergency department triage and blood sampling were performed. Time metrics were recorded by team members, who subsequently logged into the system of Medical Information Records to initiate a consult or treatment.
    2. Direct conveyance to the emergency room. The neurologist re- ceived stroke alert notifications by phone and primary diagnosis and Treatment decisions were made. The time was recorded by

Fig. 1. Picture of radiofrequency identification devices. Tag comprising wristwatch (a) and mounted reader (b).

an emergency doctor in the electronic medical records system.

• • 1. Direct conveyance to the CT(Computed Tomography) room. The CT was completed and the time of report availability was recorded by a technician. IVT was initiated, and the time was recorded by a neurologist.
    2. Conveyance to the angiography suite. The time was recorded by an interventional neurologist.

After the in-hospital stroke system was implemented, the following operational workflow for AIS patients has been established:

1. Patient arrival. The emergency department nurse initiates regis- tration and places the wristwatch on the patient. Stroke alert no- tification is automatically activated by the system and the time recording process is initiated.
2. Direct conveyance to the emergency room. The RFID reader at the gate of the emergency room identifies the radiofrequency signal of the patient’s wristwatch and the time is recorded auto- matically. Blood glucose and baseline ECG assessments are com- pleted, and the times of uploading the results are recorded.
3. Direct conveyance to the CT room. The RFID reader at the gate of the CT room records the time when the patient passes by. The time of uploading the CT report is recorded. IVT is initiated and the time point is recorded in the system.
4. Conveyance to the angiography suite. The RFID reader at the gate records the time when the patient passes by. ET is initiated, and the time of initiation is recorded in the system.
   1. Data collection

We collected data on the patients’ demographics, their medical his- tory, clinical characteristics, National Institutes of Health Stroke Scale and Modified Rankin scale (mRS) scores, and ways of reaching the hospital. Additionally, time metrics (symptom OTT, door-to-blood testing, door-to-CT scan, DNT, and door-to- endovascular puncture time [DPT]) were documented during the entire emergency manage- ment process. Moreover, information on the patients’ treatments and outcomes, and factors causing treatment delays was gathered. The pri- mary outcome was in-hospital mortality, and secondary outcomes were intracerebral or splanchnic hemorrhages, differences in NIHSS scores between hospital admission and discharge, the mRS score at dis- charge, and the length of stay (LOS). Patient hospice care as a result of deteriorating conditions was considered a hospital death. Data were collected from hospital’s electronic medical records (EMR) using a med- ical chart that we designed. A trained specialist collected the data and filled in the medical charts.

• 1. Statistical analysis

Continuous variables with normal distribution were presented as mean +- standard deviations, whereas those without normal distribu- tion were expressed as median values and inter-quartile ranges. We checked the normal distribution of the variables using the Shapiro- Wilk test. Categorical data were presented as counts or percentages. Differences in continuous variables with normal distribution between two groups were analyzed using Student’s t-test. Differences in contin- uous variables without normal distribution or count data between two groups were compared using the Mann-Whitney U test. Differences in categorical data were analyzed using the R x C Crosstab Chi-square test. An interrupted time-series analysis, using a segmented logistic re- gression model, was performed to evaluate the association between the in-hospital stroke system and the traditional stroke information system. Data were processed and analyzed using the SPSS statistical package,

version 22. A p value below 0.05 (two-tailed) was considered statisti- cally significant.

• 1. Ethical considerations

This study has been recorded by the Ethics Committee of Hunan Normal University. Because it was a retrospective analysis, no informed consent was required.

1. Results
   1. Patient baseline data

From June 2017 to December 2021, we gathered a total of 1076 AIS cases. A total of 1031 AIS patients remained after excluding 45 cases (11 in-hospital strokes, 27 patients who received IVT or ET in other hospitals prior to admission, 5 patients who received head CT scans prior to admission, and 2 patients who died within 24 h prior to commencing definitive treatments). All of the patients were grouped into two catego- ries: the pre-intervention group and the post-intervention group. No statistical differences in the demographics and medical histories of patients in the two groups were found. Both groups showed similar clin- ical characteristics apart from higher NIHSS scores on admission among patients of the post-intervention group. There was no significant differ- ence between the two groups relating to their arrival at the hospital via EMS(Table 1).

• 1. Treatments and outcomes

In the interrupted time-series analysis (Fig. 2), the rate of patients received IVT or ET increased after the implementation of the in- hospital stroke system(odds ration of IVT, 6; 95% CI, 0.48-75.34; P = 0.282. odds ration of ET, 1.5, 95% CI, 0.14-16.54; P = 0.99). As shown in Table 2, The numbers of patients received IVT or ET in the post- intervention group were significantly higher than those receiving these treatments in the pre-intervention group. Moreover, the two groups showed similar rates of intracerebral or splanchnic hemorrhage and there was no significant difference in the mortality rate between the two groups. Among the surviving patients, the LOS of those in the post-intervention group was significantly shorter than that of patients in the pre-intervention group. The post-intervention group showed sig- nificantly greater differences in NIHSS scores at admission and dis- charge compared with the pre-intervention group. There was no significant statistical difference in mRS scores at discharge between the two groups.

• 1. Time metrics

In the interrupted time-series analysis, there was a significant decrease in the odds of DNT after the beginning of the in-hospital stroke system (Aug-2019), (odds ration, 0.29; 95% CI, 0.17-0.49; P < 0.001) (Fig. 2). As shown in Table 3, a total of 186 patients were treated with IVT or IVT bridging ET, of whom 23 were in the pre-intervention group and 163 were in the post-intervention group. Compared with the pre-intervention group, the post-intervention group exhibited sig- nificantly reduced symptom onset-to-door times. Their OTT values were also lower than those of the pre-intervention group. Moreover, they had lower values for door-to-blood sampling times and door-to- CT angiography times. Consequently, there was a significant decrease in DNT in the post-intervention group compared with that in the pre- intervention group. Similarly, the DPT in the post-intervention group was lower. Thus, the number of patients receiving IVT within 60 min (DNT < 60 min) increased significantly in the post-intervention group. However, patients with OTT < 60 min did not increase

Table 1

A comparison of patients’ demographic features, clinical characteristics, and ways of reaching the hospital in the pre-intervention and post-intervention groups.

Demographic features	All patients (n = 1031)	Pre-intervention group (n = 474)	Post-intervention group (n = 557)	P value
Gender, male/female	664/367	306/168	358/199	0.924
Age (years)	64.22 +- 11.57	63.88 +- 11.71	64.52 +- 11.44	0.376
Smoke, yes/no	555/476	226/248	250/307	0.369
Hypertension, yes/no,	664	309	355	0.627
Coronary heart disease, yes/no	228	107	121	0.743
Diabetes mellitus, yes/no,	223	102	121	0.937
Cerebrovascular disease	191	82	109	0.350
Clinical characteristics
Systolic pressure (mmHg)	148.06 +- 21.43	147.98 +- 20.12	148.13 +- 22.49	0.913
Diastolic pressure (mmHg)	84.04 +- 13.46	83.24 +- 13.25	84.73 +- 13.60	0.076
Stroke localization ICA	56 (20.8%)	10 (24.4%)	46 (20.2%)	0.410
MCA	186 (69.1%)	25 (61.0%)	161 (70.6%)
BA	27 (10.0%)	6 (14.6%)	21 (9.2%)
stroke mimics Triage positive	911 (88.4%)	415 (87.6%)	496 (89.1%)	0.455
Triage negative	120 (11.6%)	59 (12.4%)	61 (10.9%)
NIHSS on admission	5 (2-11)	4 (2-9)	6 (3-12.5)	<0.001
Ways to hospital
By EMS (n,%)	341 (33.07%)	149 (31.43%)	192 (34.47%)	0.302
By their own (n,%)	656 (63.63%)	317 (65.9%)	339 (60.2%)
From other hospitals (n,%)	34 (3.29%)	8 (1.69%)	26 (4.67%)

Abbreviations: ICA, intra cranial Internal carotid artery; MCA, Middle cerebral artery; BA, Basilar artery; NIHSS, National Institute of Health Stroke Scale; EMS, emergency medical service.

significantly after optimization (<=10% of patients in the post- intervention group).

1. Discussion

The results of this retrospective analysis showed that AIS emergency management improved significantly after the stroke unit was optimized following the introduction of an in-hospital stroke system. The results showed an increased number of patients treated with IVT or ET and reduced times for door-to-blood sampling, door-to-CT angiography, DNT, and DPT, which were consistent with the time metrics stipulated in the 2018 AHA/ASA guidelines. With improvements of these critical factors, reflecting the optimized AIS emergency management, the pa- tients’ LOS was significantly reduced, and their NIHSS scores at discharge decreased significantly. There was no significant increase in the rate of intracerebral hemorrhage despite the increased proportion of patients treated with IVT or ET, which rose from 8.65% to 41.11%.

It has been widely recognized that IVT plays an important role in im- proving the clinical outcomes of AIS patients [9]. Every minute saved for administering IVT contributes 1.8 days of disability-adjusted life years [10]. Therefore, the treatment time remains a critical factor. DNT, which is a prime consideration in stroke management, can reflect patient-focusED workflow, which influences the effect of IVT [11]. In the present study, following optimization of the hospital’s emergency procedures through the installation of an in-hospital stroke system, the DNT of AIS patients was recorded at 26 (21-38) min, which was considerably lower than the time (60 min) recommended in the AHA/ ASA international guidelines [12]. The reduced DNT reported in this study can be attributed to the use of the in-hospital stroke system, which saving time for AIS patients, leading to increased the rate of treat- ment with IVT or ET.

In light of international guidelines, stroke units (including multidis- ciplinary teams) have been established in hospitals. However, there are many potential obstacles to the practical implementation of the guide- lines. Various factors, such as the Delayed arrival of the stroke specialist, occupation of the CT room, and insufficient multidisciplinary consulta- tions caused by lack of patient medical information, may affect the emergency procedure and result in delayed DNT [13]. As noted by Bulmet [14], the factors directly affecting the emergency procedure include nonavailability of the CT scanner, and insufficient or no commu- nication among healthcare professionals. Moreover, clinical trials

conducted at the University of Pittsburgh in the United States revealed that factors that could affect DNT included inadequate multidisciplinary communication and delays in obtaining initial brain images or other patient information [15].

Once the patient arrives at the hospital, the transmission of pre- notifications is essential to ensure readiness of the CT scanner or other required equipment. Furthermore, direct activation of the stroke team to minimize delays in the arrival of the stroke specialist and address the issue of insufficient multidisciplinary communication are recom- mended. Implementation of these strategies hinges on the provision of Real-time feedback on brain images or patient information. Moreover, during the pre-intervention period in the hospital under study, time metrics were recorded by the stroke team members, who logged into the medical records system to start a consult or treatment, often filling in the information later given the acute treatment priority. This process could have increased the possibility of misreported times. Therefore, an automated method of recording times is clearly of value for improving the quality of emergency procedures.

Given recent advances in IT, these shortcomings can be improved by optimizing the emergency procedure for AIS patients [16]. In one study [17], a mobile app was introduced to optimize acute stroke care by facilitating the entry of patient parameters and by sharing a thrombolysis checklist and brain images. Through pre-notifications on the patients’ movements provided to all on-call staff, significant improvements in DNT were achieved. Moreover, the University of Col- orado Hospital has designed unique and dynamic IT infrastructure, in- cluding cloud-based remote access to EMR and a picture archiving communication system. In addition, a reliable and rapid process for transferring images from the mobile stroke unit to the hospital has been developed, which could substantially improve Thrombolysis rates and reduce DNT times [18].

Similar to previous studies, the technology for sharing real-time in- formation in our study was associated with an stroke system, enabling the transmission of pre-notifications and efficient multidisciplinary communication. However, contrasting with previous studies, the time metrics in our in-hospital stroke system were automatically defined and recorded the by the RFID system. Therefore, the gaps leading to pro- longed DNT could easily be identified. Following the identification of gaps or disparities, specific interventions, such as unobstructed emer- gency to CT room aisle for door-to-CT delay, or early communication for treatment decisions, could be initiated. Furthermore, according to

Fig. 2. Sequence Diagram of IVT (a), ET (b) and DNT (c) in time-series analysis.

Abbreviations: IVT, intravenous thrombolysis; ET, endovascular therapy; DNT, door- to- needle time

the disparities data, appropriate measures could be developed to improve the quality of care, thereby improving the patient’s condition, their responses and those of their families, and hospital management. In this way, the entire in-hospital procedure could be optimized, as evidenced by the shortened times of door-to-blood routine test reporting, door-to-CT angiography, and the reduced DNT (from

118 min in the pre-intervention group to 26 min in the post- intervention group, that is, a decrease of 92 min) using the alert system. The results of this study also revealed a higher IVT treatment rate within 60 min of symptom onset among patients in the post- intervention group compared with those in the pre-intervention group, which is a likely outcome of reduced DNT, thus saving time dur- ing the therapeutic window. Moreover, since 2019, hospitals have been

committed to implementing a protocol for promoting stroke knowledge and education, resulting in higher IVT rate.

It is noteworthy that not all of the data indicated improvements. Some findings were unexpected, such as the relatively low percentage of arrivals (33.07%) at the hospital via EMS. Moreover, as indicated by the results for the post-intervention group, the delayed responses of patients or their families might be the main cause of prehospital delays. Therefore, the health department should prioritize pre-hospital screen- ing and the promotion of awareness and education for patients whose risk of stroke is high. In addition, although the in-hospital emergency procedure was optimized through the installation of the in-hospital stroke system, pre-hospital care was not connected to the in-hospital system, given security considerations relating to the hospital’s network.

Table 2 A comparison of the treatments and outcomes of the two groups.
	All patients	Pre-intervention group	Post-intervention group	P value
	(n = 1031)	(n = 474)	(n = 557)
Treatments
IVT (n/%)	163 (15.81%)	19 (4.01%)	144 (25.85%)	< 0.001
IVT bridging ET (n/%)	23 (2.23%)	4 (0.84%)	19 (3.41%)
ET (n,%)	84 (8.15%)	18 (3.80%)	66 (11.85%)
Other medicine treatments (Aspirin, Clopidogrel, etc.) (n/%)	761 (73.81%)	433 (91.35%)	328 (58.89%)
Primary outcome In-hospital mortality (hospice), yes, n(%)	57 (5.5%)	25 (5.27%)	32 (5.75%)	0.742
no, n(n%)	974 (95.5%)	449 (94.73%)	525 (94.25%)
Secondary outcome ICH or splanchnic hemorrhage, yes, n(n%)	101 (9.80%)	42 (8.86%)	59 (10.59%)	0.351
No, n(%)	930 (90.20%)	432 (91.14%)	498 (89.41%)
Heidelberg classification of ICH
1	62 (83.8%)	23 (85.2%)	39 (83.0%)	0.703
2	8 (10.8%)	3 (11.1%)	5 (10.6%)
3	4 (5.4%)	1 (3.7%)	3 (6.4%)
Hospital stay (days)	9 (7-12)	10 (8-12)	8 (6-11)	< 0.001
NIHSS score difference between admission and discharge	-1 (-4-0)	-1 (-2-0)	-2 (-5-0)	< 0.001
mRS of discharge	2 (1-4)	2 (1-4)	2 (1-4)	0.302

Abbreviations: IVT, intravenous thrombolysis; ET, endovascular therapy; ICH, Intracerebral hemorrhage; NIHSS, National Institute of Health stroke scale.

Consequently, insufficient communication of information between ambulances and hospitals could result in various problems. For exam- ple, several patients could arrive simultaneously at the hospital, and the CT room could be occupied [19]. As reminded by delayed symptom onset-to-door time in post-intervention group, there is considerable scope for improving both metrics. Specifically, there is a need for more effective pre-hospital emergency management. Furthermore, even though streamlined protocols associated with the in-hospital stroke system led to DNTs as short as 26 min in the present study, no >10% of patients were treated within 60 min, which is the recommended prime time for AIS patients to receive definitive treatment after symp- tom onset [20]. Therefore, the system should be extended to include the pre-hospital procedure to address these issues.

• 1. Limitations

The study had some limitations. First, there was selection bias associated with retrospective analysis. We conducted a single-center study at a high-level hospital, which generally admits more serious cases. Therefore, the cases at this hospital are unlikely to represent the entire cohort of AIS cases. Moreover, patients who visited the emer- gency department earlier but were not admitted may have been missed. In addition, because of higher male morbidity, gender selection bias was inevitable.

Second, despite the significantly reduced DNT found in this study, no significant difference was found in the in-hospital mortality rates of the pre- and post-optimization groups. However, the potential benefits of an in-hospital stroke system should not be ignored. It should be borne in mind that there is a low probability of the accidental occurrence of in-hospital mortality relating to AIS (14.4% reported in one study [21]

and 16% reported in another study [22]). In addition, there was no difference in the mRS scores of patients at discharge in the two groups. Therefore, the sample size should be increased in future studies, and the follow-up visiting time should be prolonged.

Third, although a stroke unit had been established, the in-hospital stroke system was not connected to the ambulance system or to other hospitals in the same region. Consequently, pre-hospital and inter- hospital patient data, which are critical requirements in the treatment of AIS patients, were missing.

Fourth, documentation of the reasons for delays was only included as a component of the in-hospital stroke system from August 2019. Therefore, data prior to this time were not available, and the reasons why patients did not receive IVT or ET treatment were not recorded. This gap needs to be addressed in future studies.

Lastly, many cases with incomplete information were excluded, which is a defect in retrospective analysis. Some of the required information was not available, such as the time when first aid was administered, rehabilitation treatment, and long-term survival. There- fore, Prospective analysis is required to address these information gaps. Moreover, the incomplete records of treatment time metrics prior to the establishment of the in-hospital stroke system impeded an in-depth analysis of the entire process of emergency management of acute stroke.

1. Conclusion

The implementation of a system for sharing real-time information ensured the transmission of pre-notifications to team members and ef- ficient multidisciplinary communication. Moreover, time metrics were automatically recorded by the RFID system, and gaps that prolonged

Table 3

Time metrics of patients treated with IVT in the pre-intervention and post-intervention groups.

	Pre-intervention group (n = 23)	Post-intervention group (n = 163)	P value
Symptom onset-to-door time (h)	1.48 (0.84-2.50)	2.52 (1.44-3.79)	0.014
Door-to-blood sampling time (min)	54.5 (40.25-80)	12 (10-17)	< 0.001
Door-to-CT time (min)	61 (41-97.75)	16 (13-20)	< 0.001
DNT (min)	118 (80.5-137)	26 (21-38)	< 0.001
DPT (min)	193.5 (150.25-217)	89 (77.5-118.25)	< 0.001
Cases with DNT < 60 min, n(%)	4 (17.39%)	151 (92.64%)	< 0.001
Cases with DNT > 60 min, n(%)	19 (82.61%)	12 (7.36%)
OTT (h)	3.68 (2.91-4.21)	3.15 (2.08-4.33)	0.049
Cases with OTT < 60 min, n(%)	2 (8.70%)	16 (9.82%)	0.865
Cases with OTT > 60 min, n(%)	21 (91.30%)	147 (90.18%)

Abbreviations: DNT, door-to-needle time; DPT, door-to-endovascular puncture time; OTT, symptom onset-to-treatment time.

DNT were easily identified. Consequently, continuous efforts to improve the quality of care could be easily implemented. Therefore, with the in- stallation of the in-hospital stroke system for optimizing emergency procedures, DNT was significantly reduced, contributing to improved outcomes, including reduced LOS and improved NIHSS scores.

Ethics statement

The studies were reviewed and approved by the Ethics Committee of Hunan Normal University. Written informed consent for participation was not required for this study as per the national legislation and the in- stitutional requirements.

Authors contributions

Yixiong Zhang designed the study methodology, compiled statistics, conducted the analysis, and wrote the first draft of the manuscript. Yimin Zhu led the study’s implementation, conducted the analysis, and reviewed the manuscript. Tao Jiang implemented the study and data collection. Jun Liu implemented the study, collected the data, and wrote sections of the manuscript. Xianyi Tang and Wenchen Yi imple- mented the study and collected data. All of the authors contributed to the article and have approved the submitted version.

CRediT authorship contribution statement

Yixiong Zhang: Writing – original draft, Methodology. Yimin Zhu: Supervision, Project administration. Tao Jiang: Writing – review & editing, Investigation. Jun Liu: Writing – review & editing, Supervision, Formal analysis, Conceptualization. Xianyi Tang: Formal analysis, Data curation. Weichen Yi: Formal analysis, Data curation.

Data availability

The raw data supporting the conclusions of this article can be requested from the authors.

Declaration of Competing Interest

The authors declare that this study was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

This study was financially supported by Hunan Provincial Health Commission (grant no. D202310008879), and the Key Project of Hunan Provincial Science and Technology Innovation (grant no. 2020 SK 1015).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2023.04.008.

References

1. Trent SA, Johnson MA, Morse EA, Haukoos JS, et al. Patient, provider, and environ- mental factors associated with adherence to cardiovascular and cerebrovascular clinical practice guidelines in the ED. Am J Emerg Med. 2018;36(8):1397-404.
2. Mboi N, Murty Surbakti I, Trihandini I, Hay SI, et al. On the road to universal health care in Indonesia, 1990-2016: a systematic analysis for the global burden of disease study 2016. Lancet. 2018;392(10147):581-91.
3. Meretoja A, Keshtkaran M, Tatlisumak T, Churilov L, et al. Endovascular therapy for ischemic stroke: save a minute-save a week. Neurology. 2017;88(22):2123-7.
4. Powers WJ, Rabinstein AA, Ackerson T, American Heart Association Stroke Council, et al. 2018 Guidelines for the Early Management of Patients With Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Associa- tion/American Stroke Association. Stroke. 2018;49(3):e46-110.
5. Berge E, Whiteley W, Audebert H, Turc G. European Stroke Organisation (ESO) guidelines on intravenous thrombolysis for acute ischaemic stroke. Eur Stroke J. 2021.;6(1) I-LXII.
6. Heo JH, Kim YD, Nam HS, Oh MS, et al. A computerized in-hospital alert system for thrombolysis in acute stroke. Stroke. 2010;41(9):1978-83. 09-01.
7. Lee JS, Finch H, Higa K, Schroeppel TJ, et al. STRAUMA: a novel alert system for a combined stroke and trauma. Am Surg. 2022. 06-30;:31348221111510.
8. Shkirkova K, Akam EY, Huang J, Saver JL, et al. Feasibility and utility of an integrated medical imaging and informatics smartphone system for management of acute stroke. Int J Stroke. 2017;12(9):953-60. 12-01.
9. Whiteley WN, Emberson J, Lees KR, et al. stroke thrombolysis Trialists’ collaboration. Risk of intracerebral haemorrhage with alteplase after acute ischaemic stroke: a sec- ondary analysis of an Individual patient data meta-analysis. Lancet Neurol. 2016;15: 925-33.
10. Pan Y, Shi G. Silver Jubilee of stroke thrombolysis with Alteplase: evolution of the therapeutic window. Front Neurol. 2021;12:593887.
11. Puy L, Lamy C, Canaple S, Godefroy O, et al. Creation of an intensive care unit and or- ganizational changes in an adult emergency department: impact on acute stroke management. Am J Emerg Med. 2017;35(5):716-9.
12. Kelly AG, Holloway RG. Guideline: the AHA/ASA made 217 recommendations for early management of acute ischemic stroke in adults. Ann Intern Med. 2018;168 (12):JC63.
13. Ospel JM, Holodinsky JK, Goyal M. Management of Acute Ischemic Stroke due to large-vessel occlusion: JACC focus seminar. J Am Coll Cardiol. 2020;75(15):1832-43.
14. Bulmer T, Volders D, Kamal N. Analysis of thrombolysis process for acute ischemic stroke in urban and Rural hospitals in Nova Scotia Canada. Front Neurol. 2021; 12:645228.
15. Aghaebrahim A, Streib C, Rangaraju S, Jadhav AP. Streamlining door to recanalization processes in endovascular stroke therapy. J Neurointerv Surg. 2017;9(4):340-5.
16. Demaerschalk BM. Telemedicine or telephone consultation in patients with acute stroke. Curr Neurol Neurosci Rep. 2011;11(1):42-51.
17. Noone ML, Moideen F, Krishna RB, Salam KA. Mobile app based strategy improves door-to-needle time in the treatment of acute ischemic stroke. J Stroke Cerebrovasc Dis. 2020;29(12):105319.
18. Schimpf B, Deanda K, Severenuk DA, Jones WJ. Integration of real-time electronic health records and wireless Technology in a Mobile Stroke Unit. J Stroke Cerebrovasc Dis. 2019;28(9):2530-6.
19. Sood R, Annoni JM, Humm AM, Medlin F. Distance neurological supervision using Telestroke does not increase door-to-needle time in acute ischemic stroke manage- ment: the experience of two regional stroke units. Front Neurol. 2021;12:616620.
20. Fassbender K, Grotta JC, Walter S, Saver JL, et al. Mobile stroke units for prehospital thrombolysis, triage, and beyond: benefits and challenges. Lancet Neurol. 2017;16 (3):227-37.
21. Ideta TR, Lim E, Nakagawa K, Koenig MA. Racial and Ethnic disparities in hospital mortality among ischemic stroke patients in Hawaiil. J Stroke Cerebrovasc Dis. 2018;27(6):1458-65.
22. Mosarrezaii A, Amiri-Nikpour MR, Dindarian S, Mohammadi H, et al. Causes of mor- tality in patients after first-ever stroke: a retrospective population-based study. Brain Behav. 2021;11(10):e2294.