sci-wiki~$open stack stroke-recovery

after a stroke

eight items the owner assembled for post-stroke recovery, and the current guideline position on all of them, which is the first thing to read

kindmixedroutemixeditems8

Cerebrolysinusually called the gold standard here; ⚠️ the CURRENT Cochrane review of 7 trials and 1,773 patients finds NO effect on death and a moderate-certainty INCREASE in non-fatal serious adverse events, and the largest single trial at 1,070 patients was neutralthe labelled intravenous route, never intranasal
Vinpocetinefour placebo-controlled trials, only ONE of them double-blind and that one at high risk of bias; the meta-analysis authors themselves say the evidence is insufficient to recommend routine use, and no trial followed anyone past three monthswith foodoptional
Pentoxifylline⚠️ NAMED IN THE 2026 STROKE GUIDELINE AS NOT RECOMMENDED, and its label contraindicates it in recent cerebral or retinal haemorrhage; the Cochrane review covers 5 trials, none since 1995, and could not analyse functional outcome at allnot without a doctoroptional
Cortexin⚠️ ZERO prospectively registered trials worldwide, and the only independent blinded head-to-head found it INDISTINGUISHABLE FROM SALINE on both function and lesion volumethe labelled parenteral routeoptional
PQQthirty-two years since the first paper proposing it for stroke and not one human stroke trial has been registered; the rodent work is neonatal, intravenous, or dosed BEFORE the insult, which is prophylaxis rather than recoverymorning, with fatsoptional
Roxadustat⚠️ REJECTED BY THE FDA IN 2021 over thrombosis, death and MACE signals and still unapproved in the United States; zero human stroke trials, and the entire brain evidence is one short rodent papernot without a doctoroptional
ARA-290a coherent tissue-protection hypothesis with four registered trials, none of them cerebrovascular, the largest at 64 people; the well-supported half of the claim is that it AVOIDS erythropoietin's thrombosis risk, which is an absence of harm rather than a benefitnot without a doctoroptional
SS-31binds cardiolipin and stabilises the inner mitochondrial membrane, shown in isolated mitochondria and rat kidney and heart; ⚠️ no brain data in vivo, three missed primary endpoints in human trials, and zero stroke trials of any kindnot without a doctoroptional

8 items; 6 structures drawn from pubchem, 2 drawn generically because pubchem has no record of the name.

why it works

Stroke recovery is the one place on this site where the honest answer is almost entirely about what the evidence does NOT support, and printing that plainly is more useful than printing a protocol.

The 2026 stroke guideline gives pharmacological neuroprotection its strongest negative rating, Class 3 No Benefit at Level A, and states that no neuroprotective treatment is recommended to improve functional outcome. Pentoxifylline is named specifically. Cerebrolysin, vinpocetine and the rest are not discussed at all, which is a different and quieter kind of absence: they are not in the conversation.

The roster is kept because the owner assembled it and because a reader searching for this deserves to find out what is actually known rather than nothing. What each item earns is a sentence saying where its evidence really stops. Six of the eight carry an optional flag, and on this page that flag means something stronger than usual.

What genuinely helps after a stroke is not on this list: time, rehabilitation, and the secondary prevention a stroke physician prescribes.

how they work together

The intended logic is a sequence: open the microcirculation, then trigger a growth-factor response, then protect the mitochondria while the tissue is vulnerable. As reasoning it is coherent, and it is worth saying that it is coherent, because the reason to be sceptical here is not that the story is silly.

The reason to be sceptical is that the same reasoning has already been tested at scale and failed. Prehospital glyceryl trinitrate is the cautionary case the guideline itself flags: vasoactive, obviously ought to improve perfusion, and it produced net HARM in this exact population.

Three items in this roster are injectables and one is an oral drug that is not approved in the United States, so the practical overlap question is not really pharmacological.

what it feels like

Recovery after a stroke is measured in months and is dominated by rehabilitation and by spontaneous recovery, which is substantial and is the thing any addition has to beat. Attribution is extremely difficult in that setting; a reader improving on any of these has no way to know whether it was the protocol.

⚠️ READ THIS BEFORE THE ROSTER. Stroke is a medical emergency and this page is not a treatment plan. The 2026 AHA/ASA guideline states that no pharmacological or non-pharmacological neuroprotective treatment is recommended to improve functional outcome after acute ischaemic stroke, at its strongest evidence level. ⚠️ PENTOXIFYLLINE IS CONTRAINDICATED IN RECENT CEREBRAL OR RETINAL HAEMORRHAGE, ON ITS OWN LABEL. Roughly one stroke in seven is haemorrhagic, and haemorrhagic transformation of an infarct is common. A protocol titled for stroke that does not separate the two is pointing some readers at a labelled contraindication, so this page separates them: if the stroke was haemorrhagic, or the type is not known, pentoxifylline is not a candidate at all. ⚠️ ROXADUSTAT WAS REJECTED BY THE FDA IN 2021 and remains unapproved in the United States. The advisory committee voted 13 to 1 against for one indication and 12 to 2 against for the other, and the safety analyses showed increases in death, major adverse cardiac events, infection, seizure and thrombosis, with deep vein thrombosis around four times control. A stroke patient is by definition a thrombosis-risk patient. ⚠️ CEREBROLYSIN CARRIES A MEASURED INCREASE IN NON-FATAL SERIOUS ADVERSE EVENTS in the current Cochrane review, roughly 2.4 times control overall and 2.9 times on the ten day schedule most protocols use. And there is no such thing as intranasal Cerebrolysin: the route has no published human data at all, and the licensed routes are intravenous and intramuscular. ⚠️ VINPOCETINE IS NOT A LAWFUL DIETARY INGREDIENT IN THE UNITED STATES, and in 2019 the FDA advised pregnant women and women who could become pregnant not to take it, after a National Toxicology Program report found reduced fetal weight and increased miscarriage in animals at blood levels comparable to a single human dose. ⚠️ BLEEDING RISK COMPOUNDS HERE AND NOBODY MENTIONS IT. Pentoxifylline, vinpocetine and astaxanthin all affect platelets, and a stroke patient is usually already on an antiplatelet or an anticoagulant. Cortexin has no registered trial anywhere and failed an independent blinded comparison against saline. SS-31 and ARA-290 are investigational with no stroke data. None of this replaces a stroke physician.

safe

nothing flaggedcurated roster

every pair here reads as compatible on the mechanisms the site holds. that is the absence of a known conflict, not a clearance.

where

5 of 8 in one basketRUPharma

RUPharmacarries 5 of 8 · Cerebrolysin · Vinpocetine · Pentoxifylline · Cortexin · Roxadustat
Kimera Chems; Code Seancode Seancarries 4 of 8 · Cerebrolysin · Roxadustat · ARA-290 · SS-31
RUO; Code Seancode Seancarries 3 of 8 · Cerebrolysin · ARA-290 · SS-31
Limitless Biochem; Code Seancode Seancarries 2 of 8 · ARA-290 · SS-31

cost

8 of 8 priced

Cerebrolysin$24.00 · $0.112/mg · RUO; Code Sean
Vinpocetine$14 · RUPharma
Pentoxifylline$12 · RUPharma
Cortexin$32 · RUPharma
PQQ$33.30 · Amazon
Roxadustat$91.99 · Kimera Chems; Code Sean
ARA-290$32.00 · $3.20/mg · Limitless Biochem; Code Sean
SS-31$38.00 · $3.80/mg · RUO; Code Sean

redundant

2 overlapssame lever, twice

Reactive oxygen speciesreceptorPQQ · SS-31
Red blood cell deformabilityreceptorVinpocetine · Pentoxifylline
an overlap is not automatically waste; a precursor and a reuptake inhibitor share a system on purpose. it is worth knowing you are paying for the same lever twice.

open in the builderthis roster, loaded into interactions + stacks compare side by sidetwo entries, their targets and their evidence, in one table

interested in protocols? join the discord; that is where dosing, timing and the practical side get discussed.

the sci-wiki publishes this as a description of what these compounds do together, not as a recommendation to take them. Nothing here is medical advice.