Neurology

Factors associated with door-in-door-out times in large vessel occlusion stroke patients undergoing endovascular therapy

a b s t r a c t

Introduction: In the management of Large vessel occlusion stroke (LVOS), patients are frequently evaluated first at a non-endovascular stroke center and later transferred to an endovascular stroke center (ESC) for endovascular treatment (EVT). The door-in-door-out time (DIDO) is frequently used as a benchmark for transferring hospitals though there is no universally accepted nor evidenced-based DIDO time. The goal of this study was to identify factors affecting DIDO times in LVOS patients who ultimately underwent EVT.

Methods: The Optimizing Prehospital use of stroke systems of Care-Reacting to Changing Paradigms (OPUS- REACH) registry is comprised of all LVOS patients who underwent EVT at one of nine endovascular centers in the Northeast United States between 2015 and 2020. We queried the registry for all patients who were transferred from a non-ESC to one of the nine ESCs for EVT. Univariate analysis was performed using t-tests to obtain a p value. A priori, we defined a p value of <0.05 as significant. Multiple logistic regression was conducted to determine the association of variables to estimate an odds ratio.

Results: 511 patients were included in the final analysis. The mean DIDO times for all patients was 137.8 min. vascular imaging and treatment at a non-certified stroke center were associated with longer DIDO times by 23 and 14 min, respectively. On multivariate analyses, the acquisition of vascular imaging was associated with 16 additional minutes spent at the non-ESC while presentation to a non-stroke certified hospital was associated with 20 additional minutes spent at the transferring hospital. The administration of intravenous thrombolysis (IVT) was associated with 15 min less spent at the non-ESC.

Discussion: Vascular imaging and non-stroke certified stroke centers were associated with longer DIDO times. Non-ESCs should integrate vascular imaging into their workflow as feasible to reduce DIDO times. Further work examining other details regarding the transfer process such as transfer via ground or air, could help further identify opportunities to improve DIDO times.

(C) 2023

  1. Introduction

* Corresponding author at: Lewis Katz School of Medicine at Temple University, 1316 West Ontario Street, 10th floor, Philadelphia, PA 19140, United States of America.

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

In the treatment of large vessel occlusion stroke (LVOS), timely endovascular treatment (EVT) is associated with better functional outcomes at 90 days [1]. Often, stroke patients present to a non- endovascular center and subsequently require transfer to an

https://doi.org/10.1016/j.ajem.2023.04.009

0735-6757/(C) 2023

Fig. 1. Enrollment diagram.

endovascular stroke center (ESC) for EVT once a LVOS is identified [2]. Door-in-door-out time (DIDO) is frequently used as a benchmark for transferring hospitals and may be related to functional outcomes after stroke [3]. The American Heart Association does not provide a specific target for the DIDO time, although previous research has shown that a median DIDO time of <60 min can be achieved at a primary stroke center [4]. Many factors affect the transfer of LVOS patients, including diagnostic steps such imaging, therapeutic interventions such as intra- venous thrombolysis, and logistics in transferring a patient to an ESC.

The goal of this study was to examine patient and system factors as- sociated with DIDO times in a consortium of nine ESCs.

  1. Methods

This investigation was a retrospective analysis of the Optimizing Pre- hospital Use of Stroke Systems of Care-Reacting to Changing Paradigms (OPUS-REACH) registry of LVOS patients. OPUS-REACH is a consortium of a group of nine ESCs in Northeast United States. The methodology of the OPUS-REACH consortium has previously been described [5]. The registry is comprised of all LVOS patients who underwent EVT at one of the nine ESCs between 2015 and 2020. The authors queried the reg- istry for all patients who were transferred from a non-ESC for EVT. All patients who were transferred from a non-ESC and received EVT for a LVOS were included in this study. Patients were excluded if their DIDO

time were not available or their stroke occurred after arrival at hospital. If the DIDO time was available but the individual patient characteristics were not available (such as vascular imaging time), we excluded them from that specific analysis only.

Non-ESCs were divided into primary stroke center (PSCs) and

non-stroke certified hospitals. In order to characterize on-ESCs as certi- fied or non-certified, we used a variety of national and state databases. First, we accessed the database of the Joint Commission (JC) and Det Norske Veritas, Inc. (DNV) healthcare to identify certified stroke centers in New York, New Jersey, Delaware, and Pennsylvania [6,7]. We also accessed the state Department of Health websites for these four states to identify additional designated stroke centers [8-11]. We used the initial date of certification from the JC and DNV as the data for the stroke center certification. If no initial date of certification was found on the JC or DNV website, we contacted the hospital to find out their date of initial stroke center certification.

Vascular imaging was defined as a computed tomography angio- gram (CTA) or magnetic resonance imaging/angiogram (MRI/MRA) ca- pable of identifying a large vessel occlusion stroke. All hospitals in our study used CTA as their vascular imaging modality.

Descriptive summary statistics are presented as means (SDs) for continuous variables and as frequencies with percentages for categori- cal variables. A two-sample t-test was used to compare continuous variables and Chi-square test was used to compare categorical variables

Table 1

Demographics and baseline characteristics.

Race

n = 511 (%)

Table 2a

Factors associated with door-in-door-out times (univariate analysis).

Variable N = 511(%) Mean DIDO time [min] (SD)

p-value

American Indian/Alaska Native

3 (0.6)

Gender

Black or African American

67 (13.1)

Female

258(50.5)

138.2 (93.1)

0.93

White

404 (79.1)

Male

253(49.5)

137.5 (80.9)

Asian/Native Hawaiian/Pacific Islander 4 (0.8)

Race (n = 480)

0.91

American Indian/Alaska Native

3(0.6)

103.3 (37.5)

Asian/Native Hawaiian/Pacific

Islander

4(0.8)

116.3 (32.3)

Other 2 (0.4)

Unknown 31 (6.1)

Hispanic ethnicity

Yes 18 (3.5)

Black or African American

67(14.0)

131.7 (65.9)

White

404(84.2)

138.7 (90.1)

Other

2(0.4)

140.5 (14.9)

Hispanic Ethnicity (n = 504)

0.69

No

486(96.4)

138.0 (88.4)

Yes

18(3.6)

119.4 (46.5)

Means of Arrival (n = 486)

0.79

EMS from home/scene

438(90.1)

136.9 (88.1)

Private transport from home/scene

48(9.9)

128.3 (60.6)

Received IVT

0.16

No

312(61.0)

142.6 (94.0)

Yes

199(39.0)

130.2 (75.3)

Was the non-ESC in the Same Hospital System as the ESC?

0.09

No

349(68.4)

141.2 (89.6)

Yes

162(31.6)

130.2 (75.3)

Vascular imaging performed at

No 486 (95.1)

Unknown 7 (1.4)

Age

Mean (SD) 70.4 (15.1)

Range 25-103

Gender

Male 253 (49.5)

Female 258 (50.5)

Initial NIHSS

Mean (SD) 16.1 (7.8)

Last Known Well to Arrival at Non-ESC (minutes)

Mean (SD) 276(325)

Last Known Well to Arrival to EVT (minutes)

Mean (SD) 505(344)

Interval from Non-Contrast Head CT to Vascular Imaging (Minutes) [n = 219]

Mean (SD) 36(51)

between good and poor outcomes. Univariate logistic regression was

performed to establish potential factors that may contribute to faster

non-ESC (n = 386)

No

167(43.3)

122.7 (93.1)

<0.001

DIDO times. A priori, we defined a p value of <0.05 as significant in

Yes

219(56.7)

136.9 (69.7)

the univariate analysis. Multiple variable logistic regression was con-

ducted to determine the association of variables to estimate an odds

Level of Transferring Hospital Not certified

68(13.3)

157.8 (105.0)

0.26

ratio of a good outcome. Statistical analyses were performed with SAS

PSC

443(86.7)

134.7 (83.9)

Statistics Software, SAS 9.4 (SAS Institute, Cary, NC).

This study was approved by the institutional review boards of all

Last known well to arrival (n = 471)

<=4 h

312(66.2)

129.6 (72.0)

0.18

nine ESCs.

>4 h

159(33.8)

139.4 (80.7)

3. Results

Table 2b

2139 patients were screened from the OPUS-REACH registry and Factors associated with door-in-door-out times (multivariate analysis).

511 patients were included in the final analysis (Fig. 1). Demographic information is listed in Table 1. Patients were initially treated at 18 non-certified stroke centers and 38 primary stroke centers.

The mean DIDO times for all patients was 137.8 min. There were no differences in DIDO times between sexes, races, or ethnicities. Perfor- mance of vascular imaging and non-certified stroke centers were associated with longer DIDO times. (Table 2a). Advanced imaging was associated with an additional 24 min at the non-ESC while being treated at a non-certified stroke center was associated with a 23-min increase in DIDO time compared to a PSC. However, neither of these times were statistically significant. We also performed an additional analysis to de- termine if there was a relationship between baseline NIHSS and DIDO time. Comparing these two variables, the Pearson’s correlation coeffi- cient was 0.034 indicating no association between baseline NIHSS and DIDO time.

On multiple variable logistic regression, performance of vascular imaging and being a non-certified stroke center were associated with longer DIDO times. (Table 2b) The performance of vascular imaging added 16 min to the DIDO time while being a non-certified stroke center added 20 min to the DIDO time. Receipt of IVT shortened the DIDO time by 15 min. However, these times were only marginally significant.

We also performed subgroup analysis for patients with DIDO times of <60, <90, and <120 min. In the subgroup of patients with DIDO times <60 min and <90 min, the performance of vascular imaging at the transferring hospital was associated with longer DIDO times. In the subgroup of patients with DIDO times <120 min, performance of

Parameter

Time difference (minutes)

Standard error

P

value

Vascular Imaging

16.2

8.9

0.069

Non-Stroke Certified

19.7

11.9

0.099

Hospital Received IVT

-15.0

7.9

0.57

vascular imaging and treatment at a non-certified stroke center were associated with longer DIDO times.

  1. Discussion

The process of transferring a patient who may benefit from EVT from a non-ESC begins with the identification of an LVOS, and thereafter depends on several independent events. The aim of this study was to identify factors that are associated with longer DIDO times in patients who underwent EVT. While the majority of variables did not reach sta- tistical significance, there remain interesting questions regarding the process of evaluating and transferring such patients.

Certified stroke centers had shorter DIDO times compared to non- certified stroke centers. This finding is not unexpected, as certified stroke centers will have specific metrics to maintain their certification. Performance of vascular imaging was associated with a 14 min longer DIDO time. However, we found a mean time of 36 min between perfor- mance of the non-contrast head CT and CT angiogram suggesting a

possible process measure that can be improved. Although the perfor- mance of vascular imaging at a non-ESC adds a time-intensive step, the information obtained when evaluating a patient for LVOS is often a deciding factor for who will undergo EVT and ensures appropriate triage of patients to an ESC. While this study does not capture patients who underwent vascular imaging and were subsequently found not to be candidates for EVT, it does suggest that a protocol that prioritizes rapid acquisition of advanced imaging, such as CTA and CTP, is impor- tant for reducing DIDO. The additional time spent at the DIDO center may save time downstream, as the transferring center will have an endovascular team assembled to receive the patient.

We were surprised to find that receipt of IVT shortened DIDO times at non-ESCs by 15 min. We did not capture whether patients received alteplase or tenecteplase, though the vast majority of the non-ESCs in this study were using alteplase. Although the administration of IVT is a time intensive process, administration also means that a patient is early in their stroke. It is reassuring to see that non-ESCs are moving more rapidly with these patients, as the time dependent benefit of EVT is greatest early after stroke [12].

We were surprised to find no relationship between baseline NIHSS and DIDO time. We hypothesized that there would be an inverse relationship between NIHSS and DIDO time (i.e. patients with larger neurologic deficits would have shorter DIDO times). In fact, the mean NIHSS was 16 in our cohort suggesting that these patients had signifi- cant neurologic deficits. However, the Pearson’s correlation coefficient was 0.034 demonstrating no association between baseline NIHSS and DIDO time.

We were also surprised to find arrival via EMS was not associated with decreased DIDO times. Previous work regarding STEMI patients has shown that Prehospital identification and a protocol for mobiliza- tion of resources significantly reduces Door-to-balloon times [13,14]. While this effect is not seen in our data regarding stroke patients, there are several important differences. With regard to STEMI, these patients can be reliably identified on a prehospital 12 lead ECG. There- fore, the majority of STEMI patients be transported directly to a facility capable of cardiac catheterization. Our study only examines patients who initially arrived at non-ESCs and were subsequently transferred for EVT, and therefore does not capture patients who were identified as potential LVOS patients by EMS personnel and were transported directly to an ESC. In fact, with regard to patients who arrived via EMS, our cohort may be biased towards those who were either too un- stable to undergo a longer transport time or had subtle or atypical pre- sentations of LVOS and were therefore brought to a non-ESC.

    1. Limitations

Our data set does not specify whether patients were transferred via ground or Air ambulance, nor how patients were moved between hospi- tals (contracted transport units versus dedicated hospital-owned transport units). We also did not analyze transfer resources nor the time interval from the request for interfacility transfer to the time the patient left the transferring hospital. We also did not record other interventions performed in the emergency department such as ad- vanced airway placement, Arterial line placement, and infusion of anti- hypertensive medications. Such treatments may also play a significant role in DIDO times. We did not have any certified Acute Stroke Ready Hospitals in our database so we cannot comment on how these hospi- tals compare to primary stroke centers or non-stroke certified hospitals. Although we included a broad swath of stroke centers in the Northeast United States, these findings may not be applicable to other areas of the United States.

  1. Conclusions

Performance of vascular imaging at the non-ESC prolonged DIDO times but DIDO times were shortened by the administration of IVT.

Further work examining other details regarding the transfer process such as transfer via ground or air, distance to ESC, and acceptance of pa- tients in hospitals within the same system versus outside systems could help further illustrate opportunities to improve DIDO times.

CRediT authorship contribution statement Alexander Kuc: Writing – review & editing, Writing – original draft,

Data curation, Conceptualization. Derek L. Isenberg: Writing – original

draft, Supervision, Project administration, Methodology, Investigation, Data curation, Conceptualization. Chadd Kraus: Writing – review & editing, Writing – original draft, Data curation, Conceptualization. Daniel Ackerman: Writing – review & editing, Investigation, Data curation, Conceptualization. Adam Sigal: Writing – review & editing, Data curation, Conceptualization. Joseph Herres: Writing – review & editing, Investigation, Data curation, Conceptualization. Ethan S. Brandler: Writing – review & editing, Data curation, Conceptualization. Derek Cooney: Writing – review & editing, Investigation, Data curation, Conceptualization. Jason T. Nomura: Writing – review & editing, Data curation, Conceptualization. Michael Mullen: review & editing, concep- tualization. Huaqing Zhao: Writing – review & editing, Methodology. Nina T. Gentile: Writing – review & editing, Supervision, Investigation, Conceptualization.

Funding

This work was supported by the Delaware Valley Chapter of the American Heart Association.

Declaration of Competing Interest

Dr. Derek L. Isenberg and Dr. Jason T. Nomura received a grant from the Delaware Valley Chapter of the American Heart Association to partially fund this work.

This data has not been presented at any conference.

Both Dr. Derek L. Isenberg and Dr. Jason T. Nomura received funding from the Delaware Valley Chapter of the American Heart Association.

Acknowledgements

We would like to acknowledge: Judy B. Shahan, BSN, MBA (Department of Emergency Medicine, Geisinger Health), Traci S. Deaner, RN, MSN (Department of Emergency Medicine, Tower Health), Kathleen A. Murphy, RN, BSN (Department of Emergency Medicine, Christiana Care), Susan Wojcik, PhD (Department of Emergency Medicine, SUNY-Upstate).

References

  1. Saver JL, Goyal M, van der Lugt A, et al. Time to treatment with endovascular Throm- bectomy and outcomes from ischemic stroke: a Meta-analysis. JAMA. Sep 27 2016; 316(12):1279-88. https://doi.org/10.1001/jama.2016.13647.
  2. Maas WJ, Lahr MMH, Buskens E, van der Zee DJ, Uyttenboogaart M, Investigators C. Pathway design for acute stroke care in the era of endovascular thrombectomy: a critical overview of optimization efforts. Stroke. 2020;51(11):3452-60. https://doi. org/10.1161/STROKEAHA.120.030392.
  3. McTaggart RA, Moldovan K, Oliver LA, et al. Door-in-door-out time at primary stroke centers may predict outcome for emergent large vessel occlusion patients. Stroke. Dec 2018;49(12):2969-74. https://doi.org/10.1161/STROKEAHA.118.021936.
  4. Choi PMC, Tsoi AH, Pope AL, et al. Door-in-door-out time of 60 minutes for stroke with emergent large vessel occlusion at a primary stroke center. Stroke. Oct 2019; 50(10):2829-34. https://doi.org/10.1161/STROKEAHA.119.025838.
  5. Isenberg DL, Henry KA, Sigal A, et al. Optimizing Prehospital Stroke Systems of Care- Reacting to Changing Paradigms (OPUS-REACH): a pragmatic registry of large vessel occlusion stroke patients to create evidence-based stroke systems of care and elim- inate disparities in access to stroke care. BMC Neurol. Apr 07, 2022;22(1):132. https://doi.org/10.1186/s12883-022-02653-x.
  6. Commission J. Organizations that Have Achieved The Gold Seal of Approval from the

Joint Commission. Accessed April 30, 2021. https://www.qualitycheck.org/; 2023.

  1. Healthcare D. Accessed June 1, 2022. https://www.dnvhealthcareportal.com/ hospitals.
  2. NYSDOH. Stroke designated centers. New York State Department of Health; 2023. https://www.health.ny.gov/diseases/cardiovascular/stroke/designation/stroke_ designated_centers.htm.
  3. Designated Stroke Center Hospitals by County. State of New Jersey Department of Health. Accessed March 30, 2022. https://www.nj.gov/health/healthcarequality/ health-care-professionals/cardiac-stroke-services/stroke-services/list.shtml.
  4. Final Hospital Destination Capability List. Delaware Department of Health. Accessed March 30, 2022. https://dhss.delaware.gov/dhss/dph/ems/files/hospitals.pdf; 2019.
  5. Recognized Stroke Centers. Pennsylvania Department of Health. Accessed March 30,

2022. https://www.health.pa.gov/topics/EMS/Pages/Recognized-Stroke-Centers. aspx%E2%80%8BRecognized%20Stroke%20Centers.

  1. Jahan R, Saver JL, Schwamm LH, et al. Association between time to treatment with endovascular reperfusion therapy and outcomes in patients with acute ischemic stroke treated in clinical practice. JAMA. Jul 16 2019;322(3):252-63. https://doi. org/10.1001/jama.2019.8286.
  2. Kobayashi A, Misumida N, Aoi S, et al. STEMI notification by EMS predicts shorter door-to-balloon time and smaller infarct size. Am J Emerg Med. Aug 2016;34(8): 1610-3. https://doi.org/10.1016/j.ajem.2016.06.022.
  3. Stowens JC, Sonnad SS, Rosenbaum RA. Using EMS dispatch to trigger STEMI alerts decreases door-to-balloon times. West J Emerg Med. May 2015;16(3):472-80. https://doi.org/10.5811/westjem.2015.4.24248.