Celery Seed-Derived Compound: A Legitimate Neuroprotectant for Acute Ischemic Stroke?

By Neal S. Parikh, MD, MS

Assistant Professor of Neurology and Neuroscience, Weill Cornell Medical College

SYNOPSIS: A Phase III double-blind, placebo-controlled, randomized controlled trial suggests that early administration of DL-3-n-butylphthalide, when given adjunctively to thrombolysis or endovascular therapy, improves functional outcomes in patients with acute ischemic stroke. Statistically significant results of well-designed analyses are tantalizing, but confidence in the findings is tempered by a lack of generalizability, an unclear mechanism of action, and trial design irregularities.

SOURCES: Wang A, Jia B, Zhang X, et al. Efficacy and safety of butylphthalide in patients with acute ischemic stroke: A randomized clinical trial. JAMA Neurol 2023;Jun 26:e231871. doi:10.1001/jamaneurol.2023.1871. [Online ahead of print].

Wilson CW. Relative recovery and identification of carbonyl compounds from celery essential oil. J Food Sci 1970;35:766-768.

The holy grail for acute stroke is the development of a neuroprotective agent. Neuroprotective agents could amplify the benefits of effective reperfusion therapies, improve recovery odds for patients who are ineligible for reperfusion therapies, and even benefit patients with other forms of ischemic brain injury.

Without endovascular treatment, patients with large-vessel occlusions face a grim prognosis. Conversely, patients treated with endovascular therapy have a high, predictable rate of recanalization and reperfusion. Thus, these patients constitute an ideal trial population in whom to test neuroprotective agents, which may depend on reperfusion and distal tissue delivery to exert their beneficial effects.

Wang and colleagues leveraged such a trial population to conduct a Phase III efficacy and safety trial of DL-3-n-butylphthalide (NBP). NBP is found in celery seed oil and is one of the compounds thought to be responsible for the distinctive flavor and aroma of celery. Numerous preclinical and early stage clinical studies suggested that NBP may improve outcomes in ischemic stroke, with preclinical studies identifying purported anti-inflammatory and antioxidant effects, for example.

Wang and colleagues conducted a double-blind, placebo-controlled trial at 59 centers across China from 2018-2022. They included adult patients with ischemic stroke receiving either thrombolysis or endovascular treatment able to receive the trial medications within six hours of symptom onset. Patients had mild to severe stroke, as reflected by a National Institutes of Health Stroke Scale (NIHSS) score of 4-25. Patients with comorbid disability (modified Rankin Scale [mRS] score > 1) or imaging
evidence of moderate-large core infarction (Alberta Stroke Program Early CT Score [ASPECTS] ≤ 6) were excluded. Additionally, based on safety concerns, patients with renal or hepatic dysfunction, coagulopathy, and cytopenia were excluded. Patients were randomized to NBP or placebo, which was administered intravenously twice daily for up to 14 days, followed by oral capsules to complete a total of 90 days of treatment. They were followed for outcomes to 90 days.

The primary outcome originally was specified as recovery of neurological deficit, but the primary outcome was changed to proportion of patients with a favorable outcome. This was defined in a non-standard fashion: mRS 0 for patients with NIHSS 4-7, mRS 0-1 for patients with NIHSS 8-14, and mRS 0-2 for patients with NIHSS of 15-25. Secondary outcomes included change in NIHSS and infarction volume at day 14. The primary safety outcome was a serious adverse event within 90 days, including prolongation of length of stay, permanent organ damage, life-threatening conditions, and death.

The researchers screened 1,236 patients and randomized 1,216, with > 80% of patients remaining in the per-protocol populations. The study population was 68% men, with a median age of 66 years. Thrombolysis was provided in 68% of patients, and approximately 30% received endovascular treatment. Randomization resulted in well balanced groups.

The key finding was that patients randomized to NBP had 1.7-fold higher odds of a favorable functional outcome. Using a more standard outcome definition of mRS 0-2, patients randomized to NBP had 1.4-fold higher odds of a favorable outcome. At 14 days, NIHSS was lower in patients randomized to NBP, with a non-significant reduction in infarction volume. There were no differences in rates of safety outcomes.

COMMENTARY

NBP appears to be an effective and safe neuroprotectant agent when administered as an adjunct to thrombolysis and endovascular treatment for acute ischemic stroke. However, several questions remain. First, this trial was conducted in China and generalizability to other populations is not known.

Second, the trial design had two important irregularities: the primary outcome was redefined during the study’s conduct and the primary outcome definition was non-standard. Third, the authors do not report on the robustness of blinding. Given that NBP may be responsible for the flavor and aroma of celery, the possibility of unblinding is important. Social desirability bias and prevailing cultural beliefs regarding natural treatments conceivably could influence trial results if there was significant unblinding.

Finally, the mechanism of NBP is not clear. Some of these concerns could be raised about a mainstay in stroke care: aspirin. Regardless, confirmation in an international multicenter trial with robust blinding is required before incorporating NBP into routine clinical care.