This multicenter, double-blind, randomized controlled trial (RCT) aimed to assess the prevention of recurrent ischemic stroke and the safety of HH333 in patients with ischemic stroke induced by small-vessel disease. It was conducted at 3 university Korean medicine hospitals in South Korea: Kyung Hee University Korean Medicine Hospital, Dongguk University Ilsan Korean Medicine Hospital, Wonkwang University Gwangju Korean Medicine Hospital. Within 90 days of onset at baseline, patients will receive HH333 or a placebo, 2 capsules (800 mg) at a time, once a day for 720 days. The clinical effect of HH333 will be evaluated by ischemic stroke recurrence, which is the primary efficacy endpoint, at baseline and on days 30, 90, 180, 270, 360, 450, 540, 630, 720, and 750 after starting the medication. Total scores on the Fatigue Severity Scale (FSS), Fatigue Assessment Scale (FAS), Korean Patient Health Questionnaire (K-PHQ-9), and the probability of each type of Pattern Identification (fire heat, dampness-phlegm, qi deficiency or yin deficiency) will be evaluated as secondary efficacy endpoints at baseline and 90, 180, 270, 360, 450, 540, 630, and 720 days after starting medication. The total scores on the Korean National Institutes of Health Stroke Scale (K-NIHSS), modified Rankin Scale (mRS), Korean Modified Barthel Index (K-mBI), and Korean Montreal Cognitive Assessment (K-MoCA) will also be measured as exploratory endpoints at the same time as the secondary efficacy endpoints. The safety of HH333 will be assessed based on the presence of adverse events at the same time as the primary efficacy endpoints. The enrollment, interventions, and assessments of the study are summarized in Table 1. A schematic of the study process is shown in Figure 1.

Eligibility criteria and other details are listed in Textbox 1.

A total of 236 patients with ischemic stroke induced by small-vessel disease will be recruited through walk-in or hospital and school bulletin boards, public transportation billboards, newsletters, and other online advertisements. Participants who wish to participate will receive the consent form and provide written consent of their own free will. Multimedia Appendix 1 presents the research consent form. Participating in the study will be decided if the researcher determines that they are suitable through the screening process. The screening items were as follows: demographic information, medical history, history of ischemic stroke, medications, physical examination, vital signs, physical measurements, laboratory findings, electrocardiogram, and evaluation of the inclusion and exclusion criteria. All data were simultaneously recorded in a case report form.

A total of 236 participants who met the criteria were assigned to the HH333 or control group at a ratio of 1:1 by a statistician independent of this clinical study, using the block randomization method with R 4.4.0 (Comprehensive R Archive Network). Patients and investigators will be blinded to prevent access to the allocation results. Patients will be blinded to the study medication they are administered because they are taking either the treatment medication or a placebo. The treatment medication and placebo will be identical in size, shape, and appearance and packaged in the same type of packaging.

For allocation concealment, an independent statistician who generates the random numbers delivers a randomization code to each clinical trial site in a sealed, impermeable envelope. The investigators at each site will open the randomization envelopes in sequence in front of the patients to assign participants, and the open envelopes will be kept separate.

HH333 or placebo capsules (one dose: 2 capsules, 800 mg) will be orally administered with water once a day on an empty stomach after waking for 720 days. HH333 is an herbal medicine manufactured by Kyungjin Pharmaceutical Co, Ltd. (Icheon, Republic of Korea). It is a capsule formulation derived from 80% ethanol extracts of Scutellariae radix, Coptidis rhizoma, Phellodendri cortex, Gardeniae fructus, and Rhei rhizoma (Table 2). The placebo was a capsule manufactured by Kyungjin Pharmaceutical Co., Ltd. (Icheon, Republic of Korea) with the same appearance as HH333.

Medication compliance will be recorded in patients taking either HH333 or placebo. The participants will be instructed to return the empty containers of study medications taken and the remaining medications at each visit to count the actual number of doses of study medications. Compliance was calculated as follows:

The clinical efficacy of HH333 in patients with ischemic stroke induced by small-vessel disease will be evaluated based on whether ischemic stroke recurrence occurs after administration of HH333. Furthermore, the effects on quality of life and fatigue will be evaluated by measuring FSS, FAS, and K-PHQ-9 scores, and changes in pattern identification probability will be investigated to provide basic data for the Korean medicine treatment of stroke. The K-NIHSS, mRS, K-mBI, and K-MoCA will be used to assess functional improvements related to ischemic stroke. All outcome measurements will be conducted through face-to-face visits.

Recurrent ischemic stroke within 750 days while taking the study medication was investigated. Recurrent ischemic stroke is defined as a new neurological finding supported by new imaging findings. New neurological findings corresponded to the sudden onset of neurological symptoms identified on neurological examinations that did not occur prior to study participation. New radiological findings are abnormal findings on brain imaging (computed tomography or magnetic resonance), which can explain new neurological findings that did not exist before study participation. The investigation will be conducted on days 30, 90, 180, 270, 360, 450, 540, 630, 720, and day 750, which is 30 days after the end of administration.

The FSS is a 9-item, self-reported screening tool developed by Krupp et al [26] that assesses fatigue over the past week on a scale of 1 to 7 per item, with higher total scores indicating higher levels of fatigue. The FAS is a 10-item, self-reported screening tool developed by Michielsen et al [27] that examines the psychometric qualities due to fatigue on a scale from 1=“Never” to 5=“Always” per item, with a higher total score indicating higher levels of the psychological impact of fatigue. The PHQ-9, developed by Kroenke et al [28] was validated in the Korean version [29]. It comprises 9 items in total. Respondents rate the frequency of symptoms on a scale of 0-3 per item, with a total score of 10 or higher indicating major depression. These secondary endpoints will assess the clinical effectiveness of HH333 on poststroke fatigue and depression. Pattern identification is a unique diagnostic system in Korean medicine, and the diagnosis of stroke is based on 4 patterns which are fire heat, Dampness-phlegm, Qi deficiency, and Yin deficiency [30]. Jung et al [31] developed an equation model that predicts the probability of each pattern identification diagnosis given variables for stroke pattern identification and is being used for Korean medicine's standardized diagnosis of patients with stroke. Using a stroke pattern identification prediction model, the change in each pattern identified after HH333 administration was collected to obtain data for stroke treatment in Korean medicine. Investigations were conducted during the administration period on days 0 (baseline), 90, 180, 270, 360, 450, 540, 630, and 720.

The NIHSS is a tool for assessing neurological deficits resulting from stroke developed by the National Institutes of Health [32] and is currently the most widely used tool for stroke severity assessment, with a Korean version validated and used [33]. It comprises 15 questions in 11 checkpoints, each of which is scored on a scale of 0 to 4, with a total score calculated, with higher scores indicating higher severity. The mRS is a tool developed by van Swieten et al to assess the degree of disability in patients with stroke [34] and is currently the most widely used stroke functional assessment scale [35]. It is rated on a scale of 0 to 6, with higher scores indicating more severe impairment. The mBI is a tool to assess independence in activities of daily living after stroke developed by Shah et al [36], and translated with validation in Korean [37]. It contains 10 questions, each rated on a scale of 0 to 5, 0 to 10, or 0 to 15, with a higher total score indicating a lower dependency. The MoCA is a tool that examines cognitive function and was originally developed to identify mild cognitive impairment [38] but is now widely used for cognitive impairment caused by stroke, and the Korean version has been validated and used [39]. It consists of visuospatial and executive (5 points), naming (2 points), attention (6 points), language (3 points), abstraction (2 points), delayed recall (5 points) and orientation (6 points). The perfect score is 30, with a higher total score indicating higher cognitive function. Investigations were conducted during the administration period on days 0 (baseline), 90, 180, 270, 360, 450, 540, 630, and 720.

At each visit, participants will be screened for adverse events, laboratory tests, urine findings, electrocardiograms, hemorrhagic events such as cerebral hemorrhage or gastrointestinal bleeding related to antiplatelet or anticoagulant use, vital signs, and physical examinations. Laboratory findings included complete blood count (red blood cells, hemoglobin, hematocrit, platelets, and white blood cells) and biochemistry (fasting blood sugar, blood urea nitrogen, creatinine, natrium, potassium, chloride, aspartate aminotransferase, alanine aminotransferase, C-reactive protein, calcium, and uric acid). Urine findings included specific gravity, pH, albumin or protein, glucose, ketone, bilirubin, occult blood, urobilinogen, nitrite, leukocytes, and human chorionic gonadotropin in women of childbearing age. Adverse events included both self-reported symptoms and those judged by the investigator and were collected until 30 days after the end of administration.

After collecting all adverse events that appeared as test findings and symptoms in each group, we compared the frequency and severity of adverse events between the groups and their relationship with the study medications. Adverse events that are not related to the study medications do not meet the criteria for dropout but allow the patient to receive the best possible care regardless of the study. If SAEs occur, the study will be stopped immediately, in part or in whole, and the study researcher will report it to the principal investigator and the institutional review board for appropriate action.

All concomitant medications except conventional treatments for ischemic stroke are prohibited. Medications that have been used for the participants’ underlying medical condition for 12 weeks prior to screening that are considered stable and unlikely to affect the interpretation of the results of this study with no change in dosage or dose will be allowed with a note of the reason, as determined by the investigator. Transient symptoms unrelated to stroke, such as cold, flu, dyspepsia, constipation, and concomitant use, will be allowed on a one-time basis at the discretion of the investigator. The use of supplements such as vitamins is not permitted. The use of vitamin K antagonists, direct thrombin inhibitors, factor X inhibitors, and heparin (except for temporary use during percutaneous coronary intervention) that may affect bleeding propensity will be strictly prohibited.

In previous studies, ischemic stroke recurrence rates of 12.5%-12.8% over 24 months with conventional antiplatelet therapy (aspirin or dipyridamole) [22] and 1.7% over the same period with HH333 [19] were reported. Therefore, the sample size was 106 patients per group and 212 patients in total at 2 years with a control group (P₂=0.125-0.128), HH333 group (P₁=0.017), a significance level (α) of 5% and a target power of 80%. Finally, a dropout rate of 10% was considered, resulting in a sample size of 118 patients per group for a total of 236 patients.

All statistical analyses were two-sided with a 5% significance level. The statistical program SAS (version 9.4 or more, SAS Institute) was used for the analysis. If at any point there are missing values or participants are dropped before the end of the study, the data will be treated as missing data. All data obtained from the participants were analyzed in 3 forms: full analysis set, per-protocol set, and safety set. Efficacy analysis will be conducted using the results of the full analysis set as the main analysis, and a per-protocol set analysis will be conducted separately to evaluate whether there are differences from the full analysis set results. Safety analysis was performed on the safety set and the safety of the study medications was evaluated. The full analysis set includes all data from participants with at least one efficacy assessment at the primary outcome after taking the study medication. The per-protocol set included participants who were eligible for full analysis and successfully completed the study. The safety set included all data from participants who received at least one dose of the study medication.

For each group, continuous data yielded the mean, SD, median, minimum, maximum, and 95% CIs, whereas categorical data yielded frequencies and percentages. To compare groups, we used a two-sample t test or Wilcoxon rank-sum test for continuous data, and a chi-square test or Fisher’s exact test for categorical data. Baseline information that was recognized as significantly different between the 2 groups was considered a covariate in the efficacy analysis to correct for heterogeneity between the treatment groups.

An adverse event is a new occurrence of a symptom after study medication administration that was not observed before administration and includes unintended consequences such as abnormal laboratory findings, symptoms, and temporary events related to the study medications. The severity of adverse events, both self-reported and non–self-reported, will be assessed by the investigator by applying Spilker’s classification as follows: mild, not requiring additional intervention, nor significantly inhibiting the normal lifestyle (function) of the participant; moderate, significantly inhibiting the normal lifestyle (function) of the participant and may need additional intervention, recovering afterward; severe, severe adverse events requiring intensive intervention; and leaving sequelae. The causal relationship between adverse events and the study medication will be assessed by the investigator using the following six-level grading scale: (1) definitely related, (2) possibly related, (3) probably related, (4) not related, (5) definitely not related, and (6) unknown.

For all adverse events and abnormal findings on examination, the symptoms, frequency, and degree of relationship with the study medication were recorded in detail for each group. The number and proportion of participants in whom adverse events occurred at least once, 95% CI, and number of occurrences within each group will be presented. Differences in proportions between groups will be tested using the chi-square test or Fisher exact test.

This study protocol was approved by the Kyung Hee University Korean Medicine Hospital institutional review board (KOMCIRB 2023-05-010-007; approved on April 19, 2024), and conducted in accordance with the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) checklist. Multimedia Appendix 2 presents the SPIRIT checklist of the study protocol [40]. Participants who wish to participate receive an explanation of the consent form and give written consent of their own free will. All participants receive cash compensation of ₩50,000 per visit. In addition, all costs related to the clinical trial medications and examinations are covered by the research team. All data from participants will be simultaneously recorded in an electronic case report form with an identification code rather than the participant’s name to protect personal information.