SLM7 Addendum
The Vagus Nerve, the Blood-Brain Barrier, and Why the Inflammaging Cascade May Have an Off-Switch
This essay was originally published using the framework’s earlier terminology (CFM, with mechanisms SCT, IH, and RDIS). The current nomenclature is the Signal Loss Model (SLM = UC + ND + PRD). See On Renaming: From Constraint Failure to Signal Loss for the terminology map.
This addendum supplements Section III of SLM7 (”Biological State First”). It does not alter the essay’s argument. It strengthens one leg of it (the inflammaging cascade) with evidence that arrived too late for the original draft and too early to ignore.
Section III argues that chronic inflammation suppresses the biological capacity for change: reducing BDNF, inhibiting neurogenesis, and closing the plasticity window that pattern-level interventions require. That argument rests on the mechanisms laid out in SLM 4. What it doesn’t address is how the inflammatory state might be reversed at the biological level, independent of the pattern work that follows.
A pilot study published in February 2024 offers a window into that question.
Lespérance et al. (2024) tracked six patients with treatment-resistant depression who received implanted vagus nerve stimulation (VNS) at the Centre Hospitalier de l’Université de Montréal. The study is small, but notable for what it measured and for how long: forty plasma inflammatory markers, assessed before implantation and again after four or more years of continuous stimulation.
Three things matter here.
First, the baseline inflammation was not subtle. These patients arrived with CCL2 levels twenty-six times those of healthy controls. CXCL8 was twenty times normal. CCL17 was twelve times. IL-7, sixteen times. These are not modest elevations. They are the peripheral signature of a system running in sustained inflammatory overdrive; the inflammaging cascade made visible in a blood panel.
Second, after years of vagus nerve stimulation, those markers dropped substantially and consistently. CXCL8 fell 86%. CCL17 fell 71%. IL-7 fell 63%. CCL2 and CCL13 both fell roughly 53%. Critically, the reductions were consistent across nearly all six patients. In a study this small, uniformity of direction matters more than the p-values (which were, as it happens, significant: ranging from 0.004 to 0.04).
Third (and this is where it gets interesting for the SLM framework) the markers that changed are not the ones most commonly associated with depression. TNF-α, IL-6, and IFN-γ, the canonical acute inflammatory markers, did not change significantly. The authors’ explanation: those markers characterize acute inflammation and are typically studied in treatment-responsive depression. Their treatment-resistant patients showed a different inflammatory profile, one dominated by chemokines involved in chronic immune cell recruitment and tissue infiltration (CCL2, CCL13, CCL17) and by molecules governing blood-brain barrier permeability (VEGF-C, sFlt-1, bFGF).
This distinction matters. What the Lespérance data suggest is not merely that inflammation goes down with vagal stimulation. It is that the kind of inflammation operating in treatment-resistant cases may be structurally different from what most depression research measures. The signature is not a fire alarm. It is smoldering damage. Leukocytes crossing a compromised blood-brain barrier, infiltrating brain tissue, sustaining neuroinflammation that outlasts any acute trigger. VNS appears to reverse this, but over years, not weeks. The temporal scale is itself informative.
The blood-brain barrier angle deserves emphasis. VEGF-C, which promotes BBB permeability (making it leakier) dropped 54%. bFGF, which protects the junction proteins that hold the barrier together, nearly quadrupled. The authors propose that VNS restores BBB integrity, reducing immune cell infiltration into the brain and breaking the self-sustaining neuroinflammatory cycle. If that interpretation holds, it identifies a concrete anatomical mechanism by which peripheral inflammation crosses into central nervous system territory: a mechanism that would explain why the inflammaging cascade, once established, is so resistant to interventions that operate exclusively at the cognitive or pattern level.
There is a parallel finding worth noting. Conway et al. (2013), using PET imaging over twelve months of VNS in treatment-resistant depression, found that patients who responded to stimulation showed increased metabolic activity in the ventral tegmental area, a brainstem region where dopamine is produced. Nonresponders did not. The VTA activation in responders appeared gradually, lagging months behind the initial cortical metabolic changes. Conway’s interpretation: VNS may eventually activate dopaminergic reward pathways, but only after a slow process of neural adaptation.
Read together, the Lespérance and Conway findings suggest that VNS may be operating across at least two of the three biological gates described in Section III: reducing the chronic inflammation that suppresses plasticity (Gate 1, neuroplasticity availability) while slowly restoring dopaminergic signaling in the reward system (Gate 2, reward system responsivity). Two gates. One intervention. But requiring years of sustained stimulation to produce its effects.
This matters for the sequencing argument. If the inflammaging cascade operates through chronic BBB compromise and immune cell infiltration and not merely through circulating cytokines, then resolving it is not a matter of a single acute intervention. It is a matter of sustained biological pressure in the right direction, over sufficient time, to allow structural repair. That is consistent with SLM’s core claim: biological state restoration precedes pattern change. But it sharpens the claim by suggesting that substrate restoration may itself require duration and consistency that most intervention models don’t account for.
A caveat on scale: the Lespérance study has six patients and no control group. The Conway study had thirteen. These are pilot-level findings. They do not prove the mechanism. What they do is identify a plausible pathway (chronic neuroinflammation via BBB compromise) that is (a) consistent with the inflammaging cascade described in SLM 4, (b) measurable in peripheral blood, and (c) responsive to an intervention that operates at the autonomic-vagal level rather than the cognitive-insight level. That convergence is enough to take seriously, even at this sample size.
The broader VNS literature adds a complication that is, from the SLM perspective, a feature rather than a bug. A meta-analysis from Thanarajah’s group (Schiweck et al., 2024) found that VNS does not consistently resolve inflammation even when it resolves depression. Conway’s RECOVER trial (the largest VNS study to date, ~500 patients) showed clinical improvement in many patients whose devices were not activated: a robust placebo response. And the Lespérance team found no statistically significant correlation between inflammatory marker reductions and MADRS depression score changes (a power limitation at n=6, but worth noting).
What does this mean? Probably that VNS is acting through multiple semi-independent pathways: inflammation reduction, dopaminergic activation, serotonin and norepinephrine modulation, BDNF upregulation. And that the clinical improvement reflects their combined effect rather than any single mechanism. In other words: the three biological gates are not a single lock with a single key. They are independent constraints, each requiring its own resolution. An intervention that opens one gate but not the others will produce partial, inconsistent results, which is exactly what the VNS literature shows.
This is the pattern SLM predicts. And it is why the sequencing principle in Section IV is not a luxury but a diagnostic necessity.
References (Addendum)
Conway, Charles R., John T. Chibnall, Marie Anne Gebara, et al. 2013. “Association of Cerebral Metabolic Activity Changes with Vagus Nerve Stimulation Antidepressant Response in Treatment-Resistant Depression.” Brain Stimulation6 (5): 788–97.
Lespérance, Paul, Véronique Desbeaumes Jodoin, David Drouin, et al. 2024. “Vagus Nerve Stimulation Modulates Inflammation in Treatment-Resistant Depression Patients: A Pilot Study.” International Journal of Molecular Sciences 25 (5): 2679.
Pincott, Jena. 2025. “The Vagus Nerve’s Mysterious Role in Mental Health Untangled.” Scientific American, January 1.
Schiweck, Carmen, Sonja Sausmekat, Tong Zhao, Leona Jacobsen, Andreas Reif, and Sharmili Edwin Thanarajah. 2024. “No Consistent Evidence for the Anti-Inflammatory Effect of Vagus Nerve Stimulation in Humans: A Systematic Review and Meta-Analysis.” Brain, Behavior, and Immunity 116 (February): 237–58.
Return to: SLM7: Choosing Interventions Without Magical Thinking.
