MAAS-MI Prep Clinical Reasoning

The grammar generates the set; it does not have to be recalled.

“Tight, band-like” names a squeezing, pressure-type insult with a deep visceral reporter — that points at the perfusion-and-oxygenation family (Family II): a myocardial demand-supply mismatch.

“When I hurry” (the H2 tempo, a demand-linked trigger) strengthens Family II and weakens the oesophageal and chest-wall candidates, which rarely track exertion.

The generated set, in rough order: myocardial ischaemia (II) first; then the structural dangers built into any chest presentation — dissection, pulmonary embolism, tamponade — which are carried whatever the likely answer; oesophageal spasm and chest-wall strain sit low, because the grammar does not support them.

The point is that you read the set off the grammar and the body’s families — you do not retrieve it whole. (Small Physiology, Ch 4, the complaint equation; Family II.)

The pattern is chosen by mechanism-load — how far the grammar already constrains the set.

The first patient has a high-signal grammar: “burning” is a specific quality-word (a mucosal, luminal insult — Family IV) and there is a clear H3, worse lying down after meals, that confirms reflux. One mechanism is strongly indicated and a single question settles it — the schema-based sweep is short here, a fast read (with experience, pattern recognition), and you still check that no danger is masked.

The second patient has a low-signal, high-danger grammar: no localising quality, an acute tempo, and the patient is unwell. Little constrains the set and the catastrophic members are live — Rule-out, danger-first, run in parallel.

Same complaint, opposite patterns — because the pattern follows the load, not the location. (Small Physiology, Ch 6, choosing the pattern.)

Abrupt onset with interscapular pain is a grammar whose structural danger is aortic dissection — a Family III (vessel and fluid balance) catastrophe.

The tempo (H2: abrupt, a switch thrown) and the location put dissection into the candidate set as a structural member, not a coincidence, so the shape is Rule-out, danger-first. The feature that moves it most: a tearing or migrating quality, and a pulse or blood-pressure difference between the arms, raise it; a purely positional pain, reproducible on palpation, lowers it toward musculoskeletal.

But notice the ordering this exercise is really about: you name and rank the catastrophe first, by structure, and only then let the likely answer settle. Danger is built into the presentation, so it goes to the top of the list before any question of likelihood — because a missed dissection is not a recoverable error. (Small Physiology, Ch 6, Rule-out; Family III.)

The fast read is earned when the grammar is classic across all four dimensions and no danger is masked.

Here: the quality (crushing, central pressure) names the myocardial demand-supply insult, Family II; the tempo (on exertion, relieved by rest, minutes rather than seconds or hours) names the demand-linked mode of failure; site and radiation fit. What you still confirm is H4, the companions that would open a danger branch: the tearing or migrating quality of dissection, the pleuritic and breathless companions of pulmonary embolism, and rest or crescendo pain of an acute coronary syndrome.

A recognition is grounded rather than surface precisely because you can still name what would break it. The single read is safe here because every dimension points the same way and each structural danger has a companion you have checked and found absent. (Small Physiology, Ch 6, Pattern Recognition; Family II.)

The read latched onto a single high-signal word — “burning” — and closed before testing the dimensions that did not fit it.

Premature closure here is a fast read — pattern recognition — that fired before its safety check. Two things were never done: H4, the companions, was not asked; and “after meals” was read as proof of reflux, when after-meal chest pain can equally be cardiac — a meal raises the heart’s demand as readily as it provokes acid.

The catching move is the one from the previous exercise: before accepting a fast read, confirm that each structural danger has an absent companion. The exertional link was never sought, so the Family II candidate was never excluded.

The lesson is not that “burning” means reflux is wrong — it is that a single confirming word does not license closing the danger branch. (Small Physiology, Ch 6, Pattern Recognition; App 3, premature closure.)

No. The surface features look like the earlier angina — crushing, central, pressure — but a grounded recognition takes the read apart on its four dimensions, and the tempo does not fit: that pain came on at rest and has lasted twenty minutes without relief. Exertional, relieved-by-rest, minutes-long was the demand-supply signature; rest pain that persists is the acute-coronary-syndrome grammar — a different mode of failure in the same family.

The look-alike fired on appearance; the dimensions rescue you.

So it moves you out of a fast pattern-recognition read into Rule-out: an acute coronary syndrome is now a live candidate that must be excluded, not recognised away. A surface recognition would have accepted “angina” on its look and mis-fired on the atypical tempo; a grounded one survives because you could still decompose it and see the single dimension that did not hold. That is the whole aim of the fast read — one you can take apart on demand. (Small Physiology, Ch 6, Pattern Recognition; Ch 8, grounded recognition; Family II.)

The most efficient branch point is the question whose answer sends you down one branch and closes the other. A strong choice here is tempo-and-trigger (H2/H3): “did this come on over minutes, or over hours to days — and what were you doing?” Sudden dyspnoea at rest with pleuritic pain pushes hard toward pulmonary embolism (Family III) and pneumothorax; dyspnoea building over days with orthopnoea and ankle oedema pushes toward cardiac failure (Family II); wheeze on exposure pushes toward airways. One question about tempo reorganises the whole set, because tempo names the mode of failure.

A good branch point is not the most obvious question — it is the one that divides the set rather than merely adding a detail to one member of it. (Small Physiology, Ch 6, Branching; H2.)

The new data do not fit the branch you are on: sudden onset, pleuritic pain, a swollen calf and a clear chest are the pulmonary-embolism grammar (Family III), not decompensation, which builds over days with orthopnoea and oedema. The recovery is not to start over blindly — it is to return to the branch point (the tempo-and-companions node) and re-enter down the PE branch, carrying forward what the cardiac branch established: the heart-failure history raises the prior for cardiac disease but does not explain this tempo.

Revision is re-entering at the node the new data address, not scrapping the whole walk. And note the trap you have just escaped: anchoring on the known diagnosis — “she has heart failure, so this is heart failure” — is the branch-level cousin of premature closure. (Small Physiology, Ch 6, Branching; App 3, anchoring; Family III.)

Each danger has a discriminating feature. Subarachnoid haemorrhage — the tempo: “was it maximal within seconds, the worst headache of your life?” A thunderclap onset is its signature; a gradual build lowers it sharply. Meningitis — the companions: fever, neck stiffness, photophobia, rash, over hours. Raised intracranial pressure — the positional and diurnal pattern: worse on waking, on bending or coughing, with vomiting, and any focal or visual change. The rule-out asks these before the tension-type or migraine answer is accepted, because each danger is a structural member of the headache set. Excluding is not asking everything — it is asking the one feature per danger that moves it most. (Small Physiology, Ch 6, Rule-out; H3/H4; Family V.)

This is rule-out run in parallel: the dangers do not queue, because each could be the one and each is time-critical.

In a single pass you gather the companions and risk-features that speak to all of them together — tempo and trigger; calf swelling, immobility and hormonal or clot risk (PE); fever, cough and sputum (pneumonia); sudden onset and asymmetric breath sounds (pneumothorax) — rather than fully working up one before touching the next. The likely benign answer is neither forgotten nor accepted: it stays on the list as a candidate, but it cannot be confirmed until the parallel dangers are covered, because a benign explanation never excludes a dangerous one.

Running the rule-out in parallel is holding several branches open under one sweep of questions, ranked by cost rather than by likelihood. (Small Physiology, Ch 6, Rule-out, in parallel; Family III.)

Yes — the obviousness is the trap. The one candidate that must be excluded before “strain” is allowed to rest is deep vein thrombosis (Family III), for which this story carries a specific risk: immobility on a long-haul flight. The benign answer is likely; the dangerous one is cheap to miss now and expensive to meet again as a pulmonary embolism a week later.

You would sweep for unilateral swelling, calf tenderness along the deep veins, warmth, a measured difference in calf circumference, and the risk profile (immobility, oestrogen, malignancy, prior clot) — and formalise the pre-test probability rather than eyeball it.

The point is not that every calf ache is a DVT; it is that a presentation carrying a live, cheap-to-miss danger earns a danger sweep before the likely answer is accepted — precisely when the benign answer looks obvious. Resting on the obvious without the sweep is premature closure wearing the mask of common sense. (Small Physiology, Ch 6, Rule-out; App 3, premature closure; Family III.)

When H1 (quality) is empty, you enter at H4 — the companions — and let them assemble a family, working the equation from the other end.

Fatigue has no insult-word to read forward, so instead of reading a mechanism you gather the accompaniments and see which family they cohere into: weight loss, night sweats and lymphadenopathy point at a malignant or infective picture (Family I); breathlessness and pallor at anaemia (Family II); cold intolerance, weight gain and dry skin at hypothyroidism (Family V, regulation); low mood, anhedonia and disturbed sleep at a mood disorder.

The undifferentiated presentation is not read front-to-back — it is entered at the companions, because they are the only dimension carrying signal. This is the schema-based sweep, entered at H4. (Small Physiology, Ch 6, schema-based entered at H4; App 3.)

With no quality and no single lead, the sweep is organised by family, asking the companions that would light each one up, ranked by danger and prevalence.

First the catastrophic and common organic families: appetite and swallowing, abdominal pain, change in bowel habit or blood (gastrointestinal malignancy, Families I and IV); cough, haemoptysis, breathlessness (lung); fever, night sweats, lymph nodes (lymphoma, tuberculosis, HIV). Then regulation: heat intolerance, palpitations and tremor (hyperthyroidism, Family V), thirst and polyuria (diabetes). Then the common non-organic causes: mood, appetite, and the social and eating history (depression, poverty, and in the older patient dementia).

The order is danger-and-prevalence first, then the quieter causes. Each positive companion recruits a family and turns the open sweep into a branch you can then walk normally; a wholly negative sweep itself narrows the field and raises the non-organic candidates.

The schema-based sweep, entered at H4, is exactly this: the companions build the family when nothing else will. (Small Physiology, Ch 6, schema-based entered at H4; App 3; Families I, II, V.)

Being right about the destination says nothing about the route until you re-walk it. A sound route to “migraine” would have read the grammar (throbbing, unilateral, photophobia, prior identical attacks — Family V, a signalling and regulation disturbance with a known illness script) and then registered the tempo, H2 — and here is the pivot — that sudden and severe is the thunderclap grammar, whose structural danger is subarachnoid haemorrhage. That branch must be opened and closed before migraine is accepted, whatever the patient’s migraine history. A sound path rules out the catastrophe even when the likely answer turns out right; a lucky path accepts migraine only because it happened to be migraine. If your written reasoning reached the diagnosis without ever opening the haemorrhage branch, you arrived at the right destination by an unsound route — and the next thunderclap migraineur is the one that route fails. (Foundation §2.3, destination versus route; Small Physiology, Ch 6, Rule-out.)

Confidence should track the completeness of the traversal, not the familiarity of the pattern.

The tension-type read (band-like, chronic, bilateral — Family V and mechanical) is plausible, but three companions that would open a raised-intracranial-pressure branch were never asked (H4): morning worsening, vomiting, and visual change. Until that danger branch has been opened and closed, ninety per cent is miscalibrated — it reports how typical the pattern feels, not how much of the set has been excluded. Well-calibrated confidence here is provisional and names its own gap: “likely tension-type, but not safe at ninety per cent until the H4 companions are checked.”

Calibration is the honest reading of how complete the route is. (Foundation §2.3; App 3, calibration.)

This is the uncoupled regime: the founding assumption — that the feel carries the mechanism — no longer holds, because the reporter is sensitised rather than the tissue.

The warning signs are all present: a grammar that will not cohere across its four dimensions, companions that fit no family, a story that shifts between visits, and repeated readings that predicted a relief which never came. The move is to stop forcing the equation: name the uncertainty honestly, keep any dangerous branch covered, and shift from the reasoning traversal toward the interview’s process skills (Scales A, B and C) and shared decision-making (Scale 3).

Forcing a mechanism onto a body that is not producing one is the error to avoid. These presentations — now often grouped as persistent physical symptoms — are common, so the skill is to suspect the regime rather than be ambushed by it. (Small Physiology, Ch 6 §5; App 3, “When the grammar does not resolve”; Löwe et al., Lancet 2024;403:2649.)

There is no single right list — but a good note names the feature, not just the cases. Trust the fast read where the grammar is classic across all four dimensions, the case is ordinary, and each structural danger has an absent companion you can check in one question.

Slow down and re-walk where the case is atypical, where the stakes are high, where the pattern fired too easily, or where a single high-signal word did the work alone (the premature-closure trap).

The feature is not the diagnosis — it is how much of the traversal you have actually completed and how costly being wrong would be.

The end-state this whole Prep aims at is a fast read you could still take apart on demand: grounded recognition, not surface recognition. (Foundation §3; Small Physiology, Ch 8; Prep · Small Physiology 8.4.)