Trial Outcomes & Findings for Deep Learning Enabled Endovascular Stroke Therapy Screening in Community Hospitals (NCT NCT05838456)
NCT ID: NCT05838456
Last Updated: 2023-06-28
Results Overview
Recruitment status
COMPLETED
Study phase
NA
Target enrollment
443 participants
Primary outcome timeframe
from the time of emergency room arrival to the time of initiation of endovascular stroke therapy (about 97 minutes)
Results posted on
2023-06-28
Participant Flow
443 were enrolled, but 200 were excluded before assignment to groups. This is a stepped wedge cluster-randomized trial with 4 clusters (4 different hospitals). In a stepped wedge fashion over 3 month intervals, the 4 clusters will initiate use of the software package (Viz.AI). Each participant was only part of the study for one single period, in other words, participants did not progress to future periods.
Unit of analysis: hospitals
Participant milestones
| Measure |
Hospital 1 - 3 Months With no Viz.AI Software, Then 12 Months With Viz.AI Software
Hospital will have 3 months with no Viz.AI software then 12 months with Viz.AI software integrated into the care pathway. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions on CT angiography (CTA) images and alerts the AIS care team.
|
Hospital 2 - 6 Months With no Viz.AI Software, Then 9 Months With Viz.AI Software
Hospital will have 6 months with no Viz.AI software then 9 months with Viz.AI software integrated into the care pathway. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions on CT angiography (CTA) images and alerts the AIS care team.
|
Hospital 3 - 9 Months With no Viz.AI Software, Then 6 Months With Viz.AI Software
Hospital will have 9 months with no Viz.AI software then 6 months with Viz.AI software integrated into the care pathway. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions on CT angiography (CTA) images and alerts the AIS care team.
|
Hospital 4 - 12 Months With no Viz.AI Software, Then 3 Months With Viz.AI Software
Hospital will have 12 months with no Viz.AI software then 3 months with Viz.AI software integrated into the care pathway. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions on CT angiography (CTA) images and alerts the AIS care team.
|
|---|---|---|---|---|
|
Step 1: Months 1-3
STARTED
|
38 1
|
8 1
|
11 1
|
23 1
|
|
Step 1: Months 1-3
COMPLETED
|
38 1
|
8 1
|
11 1
|
23 1
|
|
Step 1: Months 1-3
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
0 0
|
|
Step 2: Months 4-6
STARTED
|
9 1
|
3 1
|
4 1
|
20 1
|
|
Step 2: Months 4-6
COMPLETED
|
9 1
|
3 1
|
4 1
|
20 1
|
|
Step 2: Months 4-6
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
0 0
|
|
Step 3: Months 7-9
STARTED
|
11 1
|
2 1
|
5 1
|
19 1
|
|
Step 3: Months 7-9
COMPLETED
|
11 1
|
2 1
|
5 1
|
19 1
|
|
Step 3: Months 7-9
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
0 0
|
|
Step 4: Months 10-12
STARTED
|
7 1
|
2 1
|
5 1
|
9 1
|
|
Step 4: Months 10-12
COMPLETED
|
7 1
|
2 1
|
5 1
|
9 1
|
|
Step 4: Months 10-12
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
0 0
|
|
Step 4: Months 13-15
STARTED
|
16 1
|
7 1
|
10 1
|
34 1
|
|
Step 4: Months 13-15
COMPLETED
|
16 1
|
7 1
|
10 1
|
34 1
|
|
Step 4: Months 13-15
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
0 0
|
Reasons for withdrawal
Withdrawal data not reported
Baseline Characteristics
Deep Learning Enabled Endovascular Stroke Therapy Screening in Community Hospitals
Baseline characteristics by cohort
| Measure |
no Viz.AI Software
n=140 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
Total
n=243 Participants
Total of all reporting groups
|
|---|---|---|---|
|
Age, Continuous
|
69.5 years
n=5 Participants
|
71 years
n=7 Participants
|
70 years
n=5 Participants
|
|
Sex: Female, Male
Female
|
73 Participants
n=5 Participants
|
49 Participants
n=7 Participants
|
122 Participants
n=5 Participants
|
|
Sex: Female, Male
Male
|
67 Participants
n=5 Participants
|
54 Participants
n=7 Participants
|
121 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
Race and Ethnicity · White
|
58 Participants
n=5 Participants
|
50 Participants
n=7 Participants
|
108 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
Race and Ethnicity · Black
|
42 Participants
n=5 Participants
|
27 Participants
n=7 Participants
|
69 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
Race and Ethnicity · Hispanic
|
25 Participants
n=5 Participants
|
16 Participants
n=7 Participants
|
41 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
Race and Ethnicity · Asian
|
7 Participants
n=5 Participants
|
6 Participants
n=7 Participants
|
13 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
Race and Ethnicity · Other
|
8 Participants
n=5 Participants
|
4 Participants
n=7 Participants
|
12 Participants
n=5 Participants
|
|
Region of Enrollment
United States
|
140 Participants
n=5 Participants
|
103 Participants
n=7 Participants
|
243 Participants
n=5 Participants
|
|
Number of participants with prior stroke
|
24 Participants
n=5 Participants
|
19 Participants
n=7 Participants
|
43 Participants
n=5 Participants
|
|
Number of participants with prior transient ischemic attack (TIA)
|
11 Participants
n=5 Participants
|
5 Participants
n=7 Participants
|
16 Participants
n=5 Participants
|
|
Number of Participants with hypertension
|
107 Participants
n=5 Participants
|
75 Participants
n=7 Participants
|
182 Participants
n=5 Participants
|
|
Number of participants with hyperlipidemia
|
55 Participants
n=5 Participants
|
33 Participants
n=7 Participants
|
88 Participants
n=5 Participants
|
|
Number of participants with atrial fibrillation
|
41 Participants
n=5 Participants
|
30 Participants
n=7 Participants
|
71 Participants
n=5 Participants
|
|
Number of participants with diabetes
|
46 Participants
n=5 Participants
|
23 Participants
n=7 Participants
|
69 Participants
n=5 Participants
|
|
Number of participants with history of smoking
|
28 Participants
n=5 Participants
|
23 Participants
n=7 Participants
|
51 Participants
n=5 Participants
|
|
Number of participants with congestive heart failure
|
15 Participants
n=5 Participants
|
12 Participants
n=7 Participants
|
27 Participants
n=5 Participants
|
|
Time from last known well to time of hospital arrival
|
131.5 minutes
n=5 Participants
|
147 minutes
n=7 Participants
|
132 minutes
n=5 Participants
|
|
Score on the NIH Stroke Scale (NIHSS)
|
17 score on a scale
n=5 Participants
|
16 score on a scale
n=7 Participants
|
17 score on a scale
n=5 Participants
|
|
Score on the Alberta stroke program early CT score (ASPECTS)
|
9 score on a scale
n=5 Participants
|
10 score on a scale
n=7 Participants
|
9 score on a scale
n=5 Participants
|
|
Number of participants who received intravenous tissue plasminogen activator (tPA)
|
63 Participants
n=5 Participants
|
48 Participants
n=7 Participants
|
111 Participants
n=5 Participants
|
PRIMARY outcome
Timeframe: from the time of emergency room arrival to the time of initiation of endovascular stroke therapy (about 97 minutes)Outcome measures
| Measure |
no Viz.AI Software
n=140 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Time From Emergency Room Arrival to Initiation of Endovascular Stroke Therapy ("Door-to-groin" Time)
|
100 minutes
Interval 81.0 to 116.0
|
88 minutes
Interval 65.0 to 110.0
|
SECONDARY outcome
Timeframe: at the time of initiation of endovascular stroke therapyOutcome measures
| Measure |
no Viz.AI Software
n=140 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Number of Patients Who Received With Endovascular Stroke Therapy
|
140 Participants
|
103 Participants
|
SECONDARY outcome
Timeframe: 90 daysPopulation: mRS data were not collected for 50 in the no Viz.AI software arm and 79 in the with Viz.AI software arm.
The modified Rankin Scale (mRS) is used to assess the degree of disability or dependence in the daily activities of people who have suffered a stroke or other causes of neurological disability. The scales ranges from 0-6, as follows: 0 = No symptoms; 1 = No significant disability. Able to carry out all usual activities, despite some symptoms; 2 = Slight disability. Able to look after own affairs without assistance, but unable to carry out all previous activities; 3 = Moderate disability. Requires some help, but able to walk unassisted; 4 = Moderately severe disability. Unable to attend to own bodily needs without assistance, and unable to walk unassisted; 5 = Severe disability. Requires constant nursing care and attention, bedridden, incontinent; 6 = Dead.
Outcome measures
| Measure |
no Viz.AI Software
n=90 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=24 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Number of Patients With Good Functional Outcome Defined as Modified Rankin Score (mRS) of 0-2
|
29 Participants
|
10 Participants
|
SECONDARY outcome
Timeframe: From the time of admission to the hospital to the time of discharge (about 7 days)The number of days of inpatient hospitalization.
Outcome measures
| Measure |
no Viz.AI Software
n=140 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Hospital Length of Stay
|
7 days
Interval 4.0 to 12.0
|
6 days
Interval 3.0 to 10.0
|
SECONDARY outcome
Timeframe: From the time of admission to the hospital to the time of discharge (about 7 days)Number of participants with any intracranial hemorrhage (ICH) and symptomatic ICH (Defined by ECASS II criteria)
Outcome measures
| Measure |
no Viz.AI Software
n=140 Participants
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 Participants
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Number of Patients With Intracranial Hemorrhage (ICH)
Non-Symptomatic ICH
|
17 Participants
|
17 Participants
|
|
Number of Patients With Intracranial Hemorrhage (ICH)
Symptomatic ICH
|
7 Participants
|
2 Participants
|
Adverse Events
no Viz.AI Software
Serious events: 7 serious events
Other events: 17 other events
Deaths: 44 deaths
With Viz.AI Software
Serious events: 2 serious events
Other events: 17 other events
Deaths: 13 deaths
Serious adverse events
| Measure |
no Viz.AI Software
n=140 participants at risk
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 participants at risk
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Vascular disorders
Symptomatic intracerebral hemorrhage (ICH)
|
5.0%
7/140 • Number of events 7 • From the time of admission to the hospital to the time of discharge (about 7 days)
|
1.9%
2/103 • Number of events 2 • From the time of admission to the hospital to the time of discharge (about 7 days)
|
Other adverse events
| Measure |
no Viz.AI Software
n=140 participants at risk
Time before Viz.AI software was implemented
|
With Viz.AI Software
n=103 participants at risk
Time after Viz.AI software was implemented. Viz.AI software performs artificial intelligence-based automated detection of large vessel occlusions and alerts the AIS care team.
|
|---|---|---|
|
Vascular disorders
non-symptomatic intracerebral hemorrhage (ICH)
|
12.1%
17/140 • Number of events 17 • From the time of admission to the hospital to the time of discharge (about 7 days)
|
16.5%
17/103 • Number of events 17 • From the time of admission to the hospital to the time of discharge (about 7 days)
|
Additional Information
Sunil A. Sheth, MD
The University of Texas Health Science Center at Houston
Phone: 713-500-7897
Email: [email protected]
Results disclosure agreements
- Principal investigator is a sponsor employee
- Publication restrictions are in place