Cerebrolysin for acute ischaemic stroke
Ziganshina LE, Abakumova T, Nurkhametova D, Ivanchenko K
Cochrane Database of Systematic Reviews (2023) · n=1,773
The 2023 Cochrane review pooled seven trials and 1,773 participants and concluded with moderate-certainty evidence that Cerebrolysin probably has little to no effect on all-cause death after acute ischemic stroke — the load-bearing skeptical reading of the trial body.
This is the seventh-update Cochrane Stroke Group systematic review of Cerebrolysin (and the closely related porcine-brain hydrolysate Cortexin) in acute ischemic stroke, published in October 2023 as part of the Cochrane Database of Systematic Reviews. The review identified seven randomized controlled trials of acute-ischemic-stroke patients treated within the first 48 hours after stroke onset, totaling 1,773 participants. The intervention was Cerebrolysin or Cortexin added to standard antithrombotic and supportive therapy, compared against either standard therapy alone or standard therapy plus placebo. Primary outcomes included all-cause death and serious adverse events; secondary outcomes covered functional disability, non-fatal serious adverse events, and total adverse events.
The headline finding is that Cerebrolysin or Cortexin probably result in little to no difference in all-cause death at the latest available follow-up (relative risk 0.96, 95% CI 0.65 to 1.41, across six trials and 1,689 participants, moderate-certainty evidence). The review further concluded that Cerebrolysin probably has no beneficial effect on the total number of people with serious adverse events but probably increases the number of non-fatal serious adverse events. The included trials largely did not report on poor functional outcome defined as death or dependence — the conventional disability-status composite endpoint in the modern stroke trial literature — which the review flags as a substantive evidence gap that the manufacturer's trial program has not closed.
The review used standard Cochrane methodology: pre-registered protocol, dual independent screening, risk-of-bias assessment with GRADE certainty grading. The evidence body was assessed as moderate-certainty for the primary mortality and serious-adverse-event outcomes (downgraded for imprecision), low-certainty for total adverse events, and very low-certainty for non-death attrition. The 2023 update preserves the analytic structure of the 2020 update (Ziganshina LE et al., Cochrane Database Syst Rev 2020, CD007026.pub6) and reaches a substantively similar conclusion with refined evidence-certainty grading.
The reading framed by this review is what anchors the European Stroke Organisation and European Academy of Neurology joint guideline weak recommendation against routine use of Cerebrolysin for post-stroke cognitive impairment. The contrast with the manufacturer-aligned Bornstein et al. 2018 meta-analysis — which pooled nine RCTs (1,879 patients on the primary NIHSS analysis) and reported superiority on NIHSS at day 30 — is the structural feature the Cerebrolysin page assembles: two competent reviewer groups examining substantially overlapping trial bodies and reaching opposite conclusions about routine use.
Cochrane reviews are themselves methodological products with their own structural choices. The review's selection criteria, follow-up window definition, and trial-quality threshold each shape what enters the pooled analysis; the difference in trial count between the Cochrane review (7 trials, 1,773 patients) and the Bornstein 2018 meta-analysis (9 trials, 1,879 patients on the primary NIHSS endpoint) is partly accounted for by these methodological choices and partly by the Cochrane review's confinement to the acute-stroke window. The Cochrane reviewer team includes Russian-affiliated authors, which is mechanistically relevant because a meaningful fraction of the Cerebrolysin literature is published in Russian-language journals and the team's language access shapes which trials they can evaluate. The moderate-certainty grade for the mortality outcome is downgraded for imprecision — the confidence interval (0.65 to 1.41) is wide enough to include both a 35% mortality reduction and a 41% mortality increase, which is the basis for the "little to no effect" framing rather than a strong assertion of harm. The review does not rule out a benefit on specific functional outcomes that the included trials did not adequately measure; the structural absence of standardized 90-day disability-status reporting across the trial body is itself a finding. The review applies to acute-stroke use specifically; the gray-market application of Cerebrolysin for chronic cognitive-enhancement use sits in a fundamentally different evidence regime and is not addressed by this review. The failed peptide trials archive walks the broader pattern of trial-body interpretation across the agents whose primary endpoints have been negative at the pivotal scale.
07·Member discussion
No member discussion yet.
Member-only conversation lives here — cycle notes, practitioner commentary, pattern-matching. Be the first paying member to start the thread.