Table 3.
The candidate metabolic–microbial endotype framework, with the source population, evidence type, and directness to PCIV stated for each proposed relationship.
| Proposed level | Phenotypic features (source population) | Evidence type and directness to PCIV | Proposed acupuncture axis (evidence type) | Representative acupoints (evidence base) | Key limitations |
|---|---|---|---|---|---|
| (i) Metabolic | ↑ TC, LDL-C, TG; ↓ HDL-C; ↑ fasting glucose; ↑ HOMA-IR (phlegm-dampness constitution cohorts) (11, 30) | Human, constitution-defined, cross-sectional; indirect to PCIV | Lipid regulation via ABCA1, PPARα, LXRα; reduced foam-cell formation; hepatic IL-17 modulation (preclinical: hyperlipidemic rats, ApoE-knockout mice) (67, 68) | ST40, PC6 (animal evidence only; ST40 not reported in PCIV trials) | Features not confirmed in pattern-defined PCIV; no human evidence that acupuncture improves dyslipidemia in PCIV |
| (ii) Microbial | ↓ Flavonifractor plautii; ↓ phytosphingosine (constitution cohorts; causal data murine) (9) | Human association + mouse FMT; indirect to PCIV | ↑ Acetate, propionate, total SCFAs; microbial composition modulation (preclinical MCAO rats) (63); PROKR2-Cre vagal–adrenal axis (low-intensity electroacupuncture, mice) (12) | ST36 + LI11 (animal evidence only) | F. plautii–phytosphingosine pathway never studied after acupuncture or in PCIV; vagal–adrenal findings not generalizable to manual acupuncture |
| (iii) Inflammatory–oxidative | ↑ TNF-α–NF-κB, JAK–STAT, interferon signaling; ↓ MAIT cells (single scRNA-seq study); associated transcriptomic profiles (constitution cohorts) (10, 31) | Human, constitution-defined, cross-sectional; indirect to PCIV | ↓ NF-κB via A20/ABIN1; ↓ NLRP3 inflammasome (α7nAChR-mediated); ↓ pyroptosis (preclinical MCAO/ischemia–reperfusion) (49, 50, 54, 56); ↑ Nrf2/HO-1/NQO1, ↓ ferroptosis (preclinical vascular dementia/MCAO) (14, 58) | GV20 + LI4 + LR3; GV20 + GV14 (animal evidence only) | Anterior-circulation models; no PCIV or pattern-stratified data; MAIT finding unreplicated; oxidative phenotype not measured in phlegm-dampness cohorts |
| (iv) Vascular | ↓ Vertebrobasilar flow velocity on TCD (PCIV patients without pattern classification) (34) | Human PCIV evidence; direct to the condition, not to the pattern | ↑ BA/LVA/RVA flow velocity (human PCIV trials; healthy-volunteer TCD studies) (34, 36, 37); endothelial NO signaling (mice) (46) | GB20 ± GB12, BL10 within multi-point protocols (used in PCIV trials) (34, 35) | TCD velocity ≠ volumetric flow, perfusion, or autoregulation; the GB20 + GB12 + BL10 combination not independently evaluated; no link from velocity change to clinical outcome |
All mappings are hypotheses; none has been validated in pattern-defined PCIV patients.
TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostatic model assessment of insulin resistance; SCFA, short-chain fatty acid; PPARα, peroxisome proliferator-activated receptor α; LXRα, liver X receptor α; ABCA1, ATP-binding cassette transporter A1; MAIT, mucosal-associated invariant T cell; NF-κB, nuclear factor kappa B; ABIN1, A20-binding inhibitor of NF-κB activation 1; NLRP3, NOD-like receptor protein 3; α7nAChR, α7 nicotinic acetylcholine receptor; HO-1, heme oxygenase-1; NQO1, NAD(P)H quinone oxidoreductase 1; GPX4, glutathione peroxidase 4; SLC7A11, solute carrier family 7 member 11; TCD, transcranial Doppler; BA, basilar artery; LVA/RVA, left/right vertebral artery; MCAO, middle cerebral artery occlusion; FMT, fecal microbiota transplantation; NO, nitric oxide.