Otherwise Healthy

Two moments of institutional erasure. The biochemistry diagram in medical school, and the “otherwise healthy” participant in the clinical trial. Together they produce a body that has never been seen in the clinic, and a discipline that has agreed to see it instead.

Consider the biochemistry lecture. A diagram of the Krebs cycle occupies the board. Eight metabolites in a circle, seven enzymes on the arrows, the standard mnemonic on the students’ notes. The lecturer moves clockwise. The students copy clockwise. There will be an exam. The exam will test the order of the metabolites and the identity of the enzymes.

What the exam will not test is what the diagram assumes.

Citrate becomes isocitrate through aconitase, which requires an intact iron-sulfur cluster. The cluster requires iron reserves not depleted by chronic inflammation, and sulfur derived from methionine cycling adequately. Isocitrate becomes alpha-ketoglutarate through isocitrate dehydrogenase, which requires NAD, which requires niacin. Alpha-ketoglutarate becomes succinyl-CoA through a complex requiring thiamine, riboflavin, niacin, lipoic acid, and pantothenate at once. Every subsequent step has a similar list. Magnesium is required as a co-factor at almost every phosphorylation event. None of this is on the board.

The diagram is not incorrect. It is drawn under a scope condition that erases the class of failure that will define the students’ future patients. It draws the mechanism as if every input were supplied and every enzyme functional. In real bodies, one or several of the requirements is often absent or partial. The cycle turns, but it turns slowly, or incompletely, or at the cost of the pools upstream and downstream. The diagram is the cycle in a room with an infinite pantry. The patient will not live in that room.

The schematic body

The Krebs cycle is not exceptional. It is the shape of the whole curriculum.

Membrane transport is drawn as a channel with an arrow through it, without any drawn requirement for adequate ATP, magnesium at the ATPase, cholesterol at the correct membrane fraction, EPA and DHA at the phospholipid composition (see Omega-3), or intact glycocalyx.

Inflammation is drawn as a cascade with arrows converging on activation, without any drawn representation of the resolution phase, which is where the omega-3-derived resolvins, protectins, and maresins would appear if the class of substrate were considered as possibly absent.

Signalling pathways are drawn with clean forward arrows, without any drawn allowance for the possibility that the enzyme performing each step is undersupplied with the cofactor it requires.

Every one of these omissions is understandable inside the pedagogical frame. A diagram that included the insufficiency of every input would not fit on a slide. The exam that tested the insufficiency of every input would not scale. The pedagogical simplification is defensible. What is not defensible is that the graduate carries the simplification forward without being told they carry it, and that the profession then confirms the simplification through the evidence it accepts.

The student passes. The student becomes a physician. The physician’s internalised body is the biochemically ideal body. Every patient they meet is, for the physician’s inner eye, that body. When a symptom arrives that does not fit the body’s ideal function, the physician looks for a lesion, a mutation, a disease. What they do not look for, because the body they carry does not admit of it, is a chronic partial insufficiency at a step in a diagram that was never questioned.

The trial magic phrase

The second moment of the abstraction happens when the physician, later, is asked to trust an evidence base. The evidence base is built out of clinical trials. Clinical trials recruit participants. Participants are described, in the methods sections, by their diagnosis and by a phrase that repeats across the literature with the reliability of a liturgical formula.

Otherwise healthy.

In practice, otherwise healthy means: no diagnosed comorbidity from the exclusion list, no medication from the interaction list, no obviously alarming value on the small panel of tests actually performed at screening. It does not mean healthy. It means not documented as unhealthy on the axes the trial happened to check.

Consider a trial of a new antidepressant. It recruits a hundred and twenty adult participants with a major depressive episode. They are, per the methods section, otherwise healthy.

What is not measured at screening, and therefore not documented, and therefore not disqualifying, includes: serum 25-hydroxyvitamin D, erythrocyte magnesium, the Omega-3 Index, free triiodothyronine and reverse T3, homocysteine, methylmalonic acid, and holotranscobalamin, lipid-corrected ferritin, gamma-tocopherol (see Vitamin E), fasting insulin and HOMA-IR, the salivary cortisol curve, thyroid antibodies, vitamin K2 status, choline and betaine, potassium at the cellular level.

Any one of these being deficient would materially affect the outcome measured by the trial. Several of them being deficient at once, in a person selected for a depressive episode, is the base rate condition of the population being sampled. Depression correlates, on the aggregate, with low D, low Mg, low omega-3, low fT3 despite normal TSH, low B12 or folate in a substantial minority, and chronic low-grade inflammation. The trial does not measure. The trial does not exclude. The trial does not adjust. The trial assumes.

The assumption is that the drug acts on a physiology otherwise supplied. That assumption was inherited from an explanatory model, the monoamine hypothesis, according to which depression is a deficit of serotonergic transmission at central synapses. The model was quietly abandoned by the field in the mid-2000s in the face of accumulating counter-evidence. The trials continued. The prescriptions continued. The framework of otherwise healthy but for the serotonin continued as if the theory that justified it had not been retracted.

A generation of trial reports on antidepressants has been generated by this arrangement. The reports credit the small measured effect to the drug. The residual variance is attributed to placebo response, to intersubject variability, to sampling noise. It is not attributed to the seven other insufficiencies the trial never looked at, whose independent contribution to the symptom under study is textbook material in every field but the one prescribing.

The loop is closed by construction

The physician trained on the diagrammatic body is asked to prescribe on the basis of the trial done on the diagrammatic body. The two abstractions are the same abstraction. The one produced by the school has been reinforced by the one produced by the industry. The physician is not being asked to bridge a gap between education and evidence. They are being asked to apply an evidence base built on the assumption they were already trained to hold. Nothing pushes back.

When the patient does not respond to the treatment, the diagrammatic body remains intact. It is the patient who becomes the problem. Treatment-resistant depression enters the vocabulary. A second molecule is tried. A third. The possibility that the patient’s Krebs cycle is running on inadequate riboflavin, or that her membranes are structurally low in EPA, or that her thyroid is under-converting T4 to T3, does not arise. The mental image of the patient’s body carried by the clinician does not include that possibility. The evidence base to which the clinician defers has not been generated on a body that included it.

The loop is not open to internal critique. There is no gap in the frame at which a lever could be inserted. The frame is internally consistent. It is externally false.

The diagnostic side of the same erasure, in which the tests that would have named the deficits are almost never ordered, is developed at length in The Hunger We Don’t See. The doctrinal reply that returns the burden of proof to the discipline is developed in Rights-Based Medicine vs Evidence-Based Medicine.

The body that was not in the room

The physiological right does not apply to the body drawn on the board. It applies to the actual body in the chair across the desk. The one with three simultaneous partial deficits, a background inflammation that has been running for a decade, a membrane composition biased toward n-6 and short of n-3, and an enzyme system undersupplied at four points at once. That body is the one that came for the consultation. That body is the one that is owed correction.

The other body, the one drawn during the second year of medical school and re-enrolled at every trial screening, is nobody. It is what medicine has agreed to see instead.