Early Remote Ischemic Conditioning in Stroke (ERICS) trial

Infarct growth and recurrent stroke may be responsible for Background: early morbidity and mortality in patients with acute ischemic stroke. Remote ischemic conditioning (RIC) may reduce infarct growth and prevent recurrent stroke; however, the exact dose remains to be investigated. We hypothesized that selfor caregiver-delivered six cycles of RIC intervention in acute ischaemic stroke for the first 12 weeks is feasible and safe compared to the four cycles RIC intervention. Adult ischemic stroke patients presenting within the first 48 h of Methods: symptom onset were screened. Patients with magnetic resonance imaging (MRI) evidence of acute infarct were randomized (1:1) to receive either four or six cycles of RIC therapy sessions two times daily in both arms for 12 weeks. All patients underwent MRI for infarct volume assessment and endothelial-dependent flow-mediated dilation (EDFMD) testing at baseline, 7 days and 12 weeks. A total of 57 patients with mean±SD age of 59.4±12.4 years and Results: median National Institute of Stroke Scale, 4 (IQR, 3-7) were randomised at a median of 23 h 30 min (IQR, 10 h 20 min to 30 h) after symptom onset to either the four-cycle (n=27) or six-cycle group (n=30). A total of 18 (66%) patients completed ≥50% sessions in 12 weeks in the four-cycles group; 21 (69.7%) patients completed ≥50% sessions in 12 weeks in the six-cycle group (p=0.4). There was no between-group differences in infarct growth, early neurological deterioration, recurrent stroke, and EDFMD at 7 days and 90 days. Both four and six cycles of short-term selfor Conclusion: caregiver-delivered RIC therapy is safe and may be feasible in acute ischaemic stroke patients. Randomised clinical trials are needed to assess 1 2 3 3 4


Introduction
Early neurological deterioration (END) can occur in 14-34% of patients with acute ischemic stroke within the first few days and is associated with worse clinical outcomes 1,2 . END is attributed to progression of stroke in one third of patients, this proportion increases further in patients with single subcortical stroke 3,4 . Progression of stroke may be due to infarct growth due to persistent and progression of cerebral ischemia. Infarct growth within the first seven days independently predicts poor outcome in patients with ischemic stroke and is directly proportional to increased functional disability at 3 months as assessed by modified Rankin scale (mRS) 5 . Other mechanisms of END are re-occlusion post-recanalization, which may-occur in one-third of patients post-intravenous thrombolysis, hemorrhagic transformation, clot propagation, recurrent stroke and cerebral edema 6,7 . Furthermore, patients with large artery atherosclerosis and current smokers are more likely to have END.
Currently, strategies targeting prevention of END and recurrent stroke are early revascularization, antithrombotics, risk factor control and stroke unit management. Even in patients undergoing endovascular therapy there is increased incidence of END. Patients with persistent occlusion after endovascular therapy may have an infarct growth of up to 50 ml 8,9 . Furthermore, risk of recurrent stroke is between 4 and 8% with best medical management in the first 12 weeks 10 . Any new strategy will have to involve both the acute phase in first 7 days and up to 12 weeks post-stroke.
Remote ischaemic conditioning (RIC) involves brief cyclic ischemia (5 minutes) and reperfusion (5 minutes) of a distant organ (limb muscles) to protect at-risk (cerebral tissue) organ tissue by increasing ischemia tolerance 11 . In experimental stroke RIC alone or in combination with existing revascularization therapy may have additive effect and reduce infarct size 12,13 . Human clinical trials in acute ischemic stroke have been promising so far [14][15][16] . However, the exact mechanism of action of RIC, duration of therapy, dose of intervention, and feasibility of multiple sessions for multiple weeks in acute ischemic stroke patients are not clear. We hypothesized that self or caregiver delivered six cycles of RIC intervention in acute ischaemic stroke for the first 12 weeks is feasible and safe compared to the four-cycle RIC intervention.

Methods
Early Remote Ischemic Conditioning in Stroke (ERICS) was a prospective randomised open-label blinded-endpoint trial conducted in the Emergency and Neurology departments of a tertiary care hospital in Northwest India. The study was conducted between July 2016 and March 2018. The trial stopped after completing 3 months of follow up for all enrolled patients. The Christian Medical College and Hospital, Ludhiana Institutional Ethics Committee approved the study. (2016/IEC-0053/CMCL-Neurology). A written informed consent was obtained from all individual participants included in the study. Patients with acute transient ischaemic attack (TIA) and ischaemic stroke were block-randomised (using Research Randomizer) 1:1 to treatment with 4 or 6 cycles of RIC therapy twice daily for 12 weeks in addition to standard of care management.

Study patients
Patients presenting with first-ever ischemic stroke within first 48 h of the index event were screened. All patients ≥18 years of age, with transient ischemic attack (TIA) or mild to moderate neurological deficits (National Institute of Health Stroke Scale (NIHSS), 0-15) with magnetic resonance imaging (MRI) showing evidence of acute diffusion restriction suggestive of acute infarct and having at least one vascular risk factor (hypertension, diabetes mellitus, dyslipidemia, current smoking, coronary artery disease, symptomatic intracranial or extracranial atherosclerosis and valvular heart disease) were included. Patients with acute infarct volume >70 ml on MRI, premorbid disability (mRS>2), unstable blood pressure, symptomatic arrhythmia, upper limb injury, history of upper limb arterial atherosclerosis, increased bleeding risk, pregnant and lactating women, stroke due to vasculitis syndrome, patients with history of non-compliance, and patients with easy bruising tendencies were excluded.

Intervention procedure
The RIC therapy was initiated immediately after randomization ( Figure 1). The four-cycle therapy group consisted of four cycles of 5-minute bilateral upper limb ischemia follow by reperfusion for another 5 minutes. This was performed twice daily using a standard brachial blood pressure (BP) cuff with a pressure elevated 25 mmHg above the systolic BP to a maximum of 200 mmHg. The therapy was administered for a period of 12 weeks. Each week, patients could take two days break. Thus, per week 10 sessions were delivered and over the study period a total of 120 sessions were delivered. In the six-cycle therapy group a similar pattern was followed except six cycles of 5-minute bilateral upper limb ischemia follow by reperfusion for another 5 minutes were delivered.
The study coordinator/lead investigator delivered RIC therapy while in hospital. Family members were trained for at least three consecutive days for a total of five to six sessions by the study nurse/lead investigator. All patient were given three manual aneroid BP machines for home-therapy. After discharge, the trained family members delivered the RIC therapy at home. The caregiver maintained a daily therapy diary and the study coordinator made a weekly telephone call to the family to assess any fidelity and instrument issues.

Study procedures
All patients received standard of care management. At baseline symptom onset, NIHSS, vital signs, stroke risk factors and past medical history was recorded in a case record form. In addition, patients underwent MRI and measurement for endothelial dependent flow mediated dilation (EDFMD) at baseline, day 7 and day 90. Ankle brachial index (ABI) was measured at baseline and day 90. All patients completed a qualitative feedback to assess intervention non-compliance at the end of study (available as Extended data) 17 . The study coordinator administered the questionnaire. It had a total of six questions and took less than 3-4 minutes to complete.

MRI acquisition and analyses.
Patients were imaged on 1.5-T whole-body Signa Excite GE Medical Systems MRI scanner. MRI acquisition protocol consisted of axial T1-, T2-weighted images, fluid attenuated inversion recovery (FLAIR), diffusionweighted imaging (DWI) and MR angiography with time-offlight (MRA-TOF). All imaging data was anonymized and assigned a study number for the purpose of blinding researchers analyzing images. All planimetric analyses were performed using AnalyzePro v.1.0 software (AnalyzeDirect, Overland Park, KS, USA). Infarct volumes were measured using semiautomatic threshold-based segmentation techniques by two blinded raters (MK and SB; interclass correlation coefficient, 0.95). All sequential images were co-registered before assessment. Change in infarct volume was defined as the difference between baseline DWI lesion volume and day 7 FLAIR lesion volumes. Infarct growth was defined as increase in infarct volume of ≥0.2 ml 18 . MRA-TOF images were assessed for occlusion and stenosis of the cerebral arteries.

Endothelial dependent-flow mediated dilation (ED-FMD).
Endothelial function was assessed with ED-FMD with highresolution Philips ultrasound with L12-3 linear array transducer probe. A longitudinal section of brachial artery was assessed for peak velocity and arterial diameter. A cuff was placed above the observed arterial segment. The cuff was inflated 60 mmHg above the systolic blood pressure for a continuous period of 5 minutes. The post-deflation observation period began 30 seconds before deflation. The ultrasound probe is placed in the same brachial artery region as pre inflation. The total observation period was 150 seconds, 30 seconds prior to deflation and 2 minutes after deflation 19 . All measurements were made for both arms. The baseline assessment was done after the first RIC therapy session. ED-FMD was calculated as the percentage difference (pre inflation and post-deflation) in the arterial diameter and baseline arterial diameter.

Resting ankle brachial index (ABI).
After the first RIC therapy session resting ABI was assessed with vascular Doppler ultrasound with 8-MHz probe (VD-320, Vcomin, Shenzhen, China). Systolic BP was measured in the arm and followed by the same side leg. The measurements were done on both sides. The probe was placed over the brachial artery in the arm and dorsalis pedis artery in the foot. RABI was calculated as ratio of arm and foot systolic BP.

Outcome measures
The primary outcome measure was safety and feasibility of the 12 weeks of self/care-giver delivered RIC therapy in patients with high-risk acute ischemic stroke within 48 h of symptom onset. The feasibility was assessed on the basis of discontinuation of therapy (defined as stopping of study intervention for >3 consecutive days), completion rates of therapy (100%, 50-99% and <50%) and fidelity of the therapy. Qualitative feedback was taken from all patients to assess the reasons for continuation and discontinuation for therapy.
The secondary imaging outcome measures was absolute infarct volume change ≥0.2 ml at day 7 and silent infarcts on follow-up MRI. The secondary endothelial function outcome measure was percentage difference in the affected side ED-FMD at day 7 and day 90, and absolute difference between the four-and six-cycle group. The secondary clinical outcome was occurrence of early neurological deterioration (NIHSS ≥ 2) at day 7, recurrent stroke within 90 days and mRS at 90 days.

Sample size and statistical analysis
As this was a safety and feasibility study a convenience sample size of n = 50 was planned. Furthermore, a 15% drop rate was estimated, thus an additional seven patients were enrolled.
Variables with continuous data were assessed for normality using the Kolmogorov-Smirnov test. Symptom onsets to intervention duration, NIHSS, Infarct volume, mRS ED-FMD values and therapy sessions completed at 12 weeks were not normally distributed. The between-group differences were assessed using Mann-Whitney U-test. Age, BP, body mass index, and resting ABI was normally distributed and between groups differences for these was assessed using Student's t-test. The between group differences for the nominal variables was assessed with Chi-square test with Yates correction. Completion rates were expressed as percentages. The qualitative feedback was tabulated as absolute numbers. All analysis was done using SPSS version 24 (IBM, Armonk, USA).

Patient characteristics
A total of 57 patients with mean±SD age of 59.4±12.4years and median NIHSS score of 4 (IQR, 3-7) were randomised at a median of 23 h 30 min (IQR, 10 h 20 min to 30 h) after symptom onset to either the four-cycle (n=27) or six-cycle (n=30) group (Figure 2). The baseline demographic, clinical and imaging patient characteristics were well matched in both groups (Table 1).

Safety and feasibility
None of the patients developed any serious adverse events related to the study intervention during the 12 weeks study. A total of 19 (33.3%) patients discontinued the study intervention during the study period (Table 2). In total, five (8.8%) patients felt the study intervention duration was too long, three (5.2%) patients developed transient petechial rash over the forearm distal to the compression due to capillary fragility and two (3.5%) patients stopped due to pain in the arm during intervention. A further five (8.8%) patients were not able to take therapy for two times daily in both hands in each group. Three (5.2%) patients had thrombophlebitis in the cubital vein and in two patients the post stroke-paresthesia worsened during the therapy in both the groups. None of the patients discontinued the therapy in the first week, except for those who developed transient petechial rash. There were ten (17.5%) patients lost-to follow-up.

Infarct volume and clinical outcomes
There was no difference in the infarct volume change between groups at day 7 and day 90 (Table 3). Nine (33.3%) patients in the four-cycle group and 13 (43.3%) in the six-cycle group developed infarct growth by day 7. An additional five (18.5%) patients in the four-cycle group and two (6.7%) patients in the six-cycle group developed infarct growth by day 90. There was no significant difference between groups.

ED-FMD and ABI
The median ED-FMD in the affected and un-affected sides was similar between groups at the baseline, day 7 and day 90 (Table 3). Also on repeated measures of ANOVA as well there was no between group difference at different time points. The baseline ABI in right and left side was similar in both groups. The ABI at 90 days in right and left side was similar in both groups (Table 3).

Discussion
Our study shows that self or caregiver delivered therapy is feasible and safe in both the four-and six-cycle groups in patients with mild to moderate acute ischemic stroke patients. Two-thirds of patients completed at least 60 intervention sessions in 6 weeks in both the groups. The primary reason for discontinuation of therapy in both groups was longer duration of intervention. The END and infarct growth was similar in both groups. There was also no between-group difference in the ED-FMD and ABI.  21 for 300 days in both arms two times daily with five cycles session in patients with intracranial atherosclerosis with an automated machine. Self/caregiver delivered manual RIC therapy in both arms may be feasible for a shorter duration (6 weeks) to prevent recurrent stroke and   infarct growth. It is possible that longer compliance could be achieved with automated device compared to the manual device.
Results using experimental animal models for acute ischemic stroke have suggested that even single five-cycle session may reduce infarct size and improve cerebral perfusion 22 . This effect may be additive to thrombolysis as well. In a single human trial, a five-cycle session twice daily in both upper limbs was able to reduce the risk of recurrent stroke and improve cerebral perfusion 21 . However, it is not known if the benefit would additive if we increase the number of cycles per session. We tested both four-and six-cycle sessions in our study. Though the study was not powered to assess difference in infarct volumes and recurrent stroke, there was no trend of difference in both groups. Furthermore, there was no difference in endothelial function as well given that RIC is an systemic protection strategy. This may indicate that six cycles gives similar results to four cycles.
Ongoing clinical trials with RIC therapy in acute ischemic stroke are delivering therapy for one session in three studies and for 3 to 7 days in three other studies 11 . In our study, 12% of patients had infarct growth even after 7 days and 10% of patients developed silent brain infarcts during the 12-week period. Silent brain infarcts may be associated with increased risk of recurrent stroke and cognitive decline 23 . Furthermore, risk of recurrent stroke is high in the first 90 days after stroke, and risk is persistent even at 1 year 24 . RIC therapy may have to be given for >7 days after acute ischemic stroke in a high risk population.
Our study has several limitations. We enrolled patients at a median of 24 h after symptom onset, it is possible that the infarct growth may be less during this period. A total of 10 (16%) patients were lost to follow-up during the study. The reason for this could be out-pocket expense involved in the standard treatment of stroke. We relied on the patient/caregiver completed diaries for the completion rates of RIC therapy. This step was taken to make it easily implementable and generalizable. We used a manual device to deliver RIC therapy, which may reduce the treatment fidelity; however, all patients were given timers and adequate training during hospital stay. We also had no sham group so we could not compare the two different intensities of RIC to a sham control group.

Conclusion
Short-term self-or caregiver-delivered manual RIC therapy is safe and feasible in acute ischaemic stroke patients. A larger randomised controlled trial is needed to prove the efficacy of RIC therapy to reduce infarct growth and recurrent stroke.

8.
Even though no serious adverse events were observed, I do not believe the conclusion about the safety of RIC is appropriate for this study design (i.e., two intervention groups) and small sample size. The safety data should be interpreted as exploratory.
Under "Methods" --> "Outcome measures", there are no specific definitions of safety measures. The results mention "serious adverse events" but there is no apparent list of what were considered and measured as SAEs.
The interpretation of differences between secondary outcomes appear to heavily rely on statistical differences. As the authors noted, this was a small study and not adequately powered to statistically detect meaningful differences. I suggest the authors revise their interpretation of secondary outcome results to focus on clinical significance and state as exploratory conclusions for replication in larger studies.
Were data collected on who was providing the intervention (self and caregiver)? Perhaps in the daily diary or weekly calls? A comparison of compliance between self and caregiver administration would be very informative for planning a full-scale RCT.
More discussion of the RIC regimen used in this study (i.e., bilateral, upper arm) in comparison to previous and ongoing studies is important. Some stroke studies administer RIC in only one (nonparetic) arm. Others administer RIC with thigh cuff inflations. How might the feasibility and compliance observed in this study depend on the regimen used and translate to others?

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Partly I am the PI of a pilot study of the feasibility of prehospital delivery of remote Competing Interests: ischemic conditioning by emergency medical services in chest pain patients (clinicaltrials.gov NCT03400579). This study is funded by the North Carolina Translational and Clinical Sciences Institute, University of North Carolina at Chapel Hill, which is supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant Award Number UL1TR001111.
Reviewer Expertise: emergency medicine, stroke, statistics I confirm that I have read this submission and believe that I have an appropriate level of